Merge pull request #1577 from eyaltoledano/next

.changeset/adopt-modifyjson-pattern.md

@@ -0,0 +1,9 @@
+---
+"task-master-ai": patch
+---
+
+Improve concurrency safety by adopting modifyJson pattern in file-storage
+
+- Refactor saveTasks, createTag, deleteTag, renameTag to use modifyJson for atomic read-modify-write operations
+- This prevents lost updates when multiple processes concurrently modify tasks.json
+- Complements the cross-process file locking added in PR #1566

.changeset/fix-file-locking.md

@@ -0,0 +1,13 @@
+---
+"task-master-ai": patch
+---
+
+Fix race condition when multiple Claude Code windows write to tasks.json simultaneously
+
+- Add cross-process file locking to prevent concurrent write collisions
+- Implement atomic writes using temp file + rename pattern to prevent partial writes
+- Re-read file inside lock to get current state, preventing lost updates from stale snapshots
+- Add stale lock detection and automatic cleanup (10-second timeout)
+- Export `withFileLock` and `withFileLockSync` utilities for use by other modules
+
+This fix prevents data loss that could occur when multiple Task Master instances (e.g., multiple Claude Code windows) access the same tasks.json file concurrently.

.changeset/list-blocks-ready-filter.md

@@ -0,0 +1,16 @@
+---
+"task-master-ai": minor
+---
+
+Add --ready and --blocking filters to list command for identifying parallelizable tasks
+
+- Add `--ready` filter to show only tasks with satisfied dependencies (ready to work on)
+- Add `--blocking` filter to show only tasks that block other tasks
+- Combine `--ready --blocking` to find high-impact tasks (ready AND blocking others)
+- Add "Blocks" column to task table showing which tasks depend on each task
+- Blocks field included in JSON output for programmatic access
+- Add "Ready" column to `tags` command showing count of ready tasks per tag
+- Add `--ready` filter to `tags` command to show only tags with available work
+- Excludes deferred/blocked tasks from ready count (only actionable statuses)
+- Add `--all-tags` option to list ready tasks across all tags (use with `--ready`)
+- Tag column shown as first column when using `--all-tags` for easy scanning

.changeset/loop-error-handling.md

@@ -0,0 +1,10 @@
+---
+"task-master-ai": "patch"
+---
+
+Improve loop command error handling and use dangerously-skip-permissions
+
+- Add proper spawn error handling (ENOENT, EACCES) with actionable messages
+- Return error info from checkSandboxAuth and runInteractiveAuth instead of silent failures
+- Use --dangerously-skip-permissions for unattended loop execution
+- Fix null exit code masking issue

.changeset/loop-sandbox-optional.md

@@ -0,0 +1,9 @@
+---
+"task-master-ai": patch
+---
+
+Make Docker sandbox mode opt-in for loop command
+
+- Add `--sandbox` flag to `task-master loop` (default: use plain `claude -p`)
+- Preserve progress.txt between runs (append instead of overwrite)
+- Display execution mode in loop startup output

.changeset/update-codex-cli-models.md

@@ -0,0 +1,10 @@
+---
+"task-master-ai": patch
+---
+
+Update Codex CLI supported models to match current available models
+
+- Remove deprecated models: gpt-5, gpt-5-codex, gpt-5.1
+- Add gpt-5.2-codex as the current default model
+- Add gpt-5.1-codex-mini for faster, cheaper option
+- Keep gpt-5.1-codex-max and gpt-5.2

apps/cli/package.json

@@ -16,8 +16,8 @@
 		"test": "vitest run",
 		"test:watch": "vitest",
 		"test:coverage": "vitest run --coverage",
-		"test:unit": "vitest run -t unit",
-		"test:integration": "vitest run -t integration",
+		"test:unit": "vitest run '**/*.spec.ts'",
+		"test:integration": "vitest run '**/*.test.ts'",
 		"test:e2e": "vitest run --dir tests/e2e",
 		"test:ci": "vitest run --coverage --reporter=dot"
 	},

apps/cli/src/commands/list.command.spec.ts

@@ -0,0 +1,959 @@
+/**
+ * @fileoverview Unit tests for ListTasksCommand
+ */
+
+import type { TmCore } from '@tm/core';
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
+
+// Mock dependencies - use importOriginal to preserve real implementations
+// Only mock createTmCore since we inject a mock tmCore directly in tests
+vi.mock('@tm/core', async (importOriginal) => {
+	const actual = await importOriginal<typeof import('@tm/core')>();
+	return {
+		...actual,
+		createTmCore: vi.fn()
+	};
+});
+
+vi.mock('../utils/project-root.js', () => ({
+	getProjectRoot: vi.fn((path?: string) => path || '/test/project')
+}));
+
+vi.mock('../utils/error-handler.js', () => ({
+	displayError: vi.fn()
+}));
+
+vi.mock('../utils/display-helpers.js', () => ({
+	displayCommandHeader: vi.fn()
+}));
+
+vi.mock('../ui/index.js', () => ({
+	calculateDependencyStatistics: vi.fn(() => ({ total: 0, blocked: 0 })),
+	calculateSubtaskStatistics: vi.fn(() => ({ total: 0, completed: 0 })),
+	calculateTaskStatistics: vi.fn(() => ({ total: 0, completed: 0 })),
+	displayDashboards: vi.fn(),
+	displayRecommendedNextTask: vi.fn(),
+	displaySuggestedNextSteps: vi.fn(),
+	getPriorityBreakdown: vi.fn(() => ({})),
+	getTaskDescription: vi.fn(() => 'Test description')
+}));
+
+vi.mock('../utils/ui.js', () => ({
+	createTaskTable: vi.fn(() => 'Table output'),
+	displayWarning: vi.fn()
+}));
+
+import { ListTasksCommand } from './list.command.js';
+
+describe('ListTasksCommand', () => {
+	let consoleLogSpy: any;
+	let mockTmCore: Partial<TmCore>;
+
+	beforeEach(() => {
+		consoleLogSpy = vi.spyOn(console, 'log').mockImplementation(() => {});
+
+		mockTmCore = {
+			tasks: {
+				list: vi.fn().mockResolvedValue({
+					tasks: [{ id: '1', title: 'Test Task', status: 'pending' }],
+					total: 1,
+					filtered: 1,
+					storageType: 'json'
+				}),
+				getStorageType: vi.fn().mockReturnValue('json')
+			} as any,
+			config: {
+				getActiveTag: vi.fn().mockReturnValue('master')
+			} as any
+		};
+	});
+
+	afterEach(() => {
+		vi.clearAllMocks();
+		consoleLogSpy.mockRestore();
+	});
+
+	describe('JSON output format', () => {
+		it('should use JSON format when --json flag is set', async () => {
+			const command = new ListTasksCommand();
+
+			// Mock the tmCore initialization
+			(command as any).tmCore = mockTmCore;
+
+			// Execute with --json flag
+			await (command as any).executeCommand({
+				json: true,
+				format: 'text' // Should be overridden by --json
+			});
+
+			// Verify JSON output was called
+			expect(consoleLogSpy).toHaveBeenCalled();
+			const output = consoleLogSpy.mock.calls[0][0];
+
+			// Should be valid JSON
+			expect(() => JSON.parse(output)).not.toThrow();
+
+			const parsed = JSON.parse(output);
+			expect(parsed).toHaveProperty('tasks');
+			expect(parsed).toHaveProperty('metadata');
+		});
+
+		it('should override --format when --json is set', async () => {
+			const command = new ListTasksCommand();
+			(command as any).tmCore = mockTmCore;
+
+			await (command as any).executeCommand({
+				json: true,
+				format: 'compact' // Should be overridden
+			});
+
+			// Should output JSON, not compact format
+			const output = consoleLogSpy.mock.calls[0][0];
+			expect(() => JSON.parse(output)).not.toThrow();
+		});
+
+		it('should use specified format when --json is not set', async () => {
+			const command = new ListTasksCommand();
+			(command as any).tmCore = mockTmCore;
+
+			await (command as any).executeCommand({
+				format: 'compact'
+			});
+
+			// Should use compact format (not JSON)
+			const output = consoleLogSpy.mock.calls;
+			// In compact mode, output is not JSON
+			expect(output.length).toBeGreaterThan(0);
+		});
+
+		it('should default to text format when neither flag is set', async () => {
+			const command = new ListTasksCommand();
+			(command as any).tmCore = mockTmCore;
+
+			await (command as any).executeCommand({});
+
+			// Should use text format (not JSON)
+			// If any console.log was called, verify it's not JSON
+			if (consoleLogSpy.mock.calls.length > 0) {
+				const output = consoleLogSpy.mock.calls[0][0];
+				// Text format output should not be parseable JSON
+				// or should be the table string we mocked
+				expect(
+					output === 'Table output' ||
+						(() => {
+							try {
+								JSON.parse(output);
+								return false;
+							} catch {
+								return true;
+							}
+						})()
+				).toBe(true);
+			}
+		});
+	});
+
+	describe('format validation', () => {
+		it('should accept valid formats', () => {
+			const command = new ListTasksCommand();
+
+			expect((command as any).validateOptions({ format: 'text' })).toBe(true);
+			expect((command as any).validateOptions({ format: 'json' })).toBe(true);
+			expect((command as any).validateOptions({ format: 'compact' })).toBe(
+				true
+			);
+		});
+
+		it('should reject invalid formats', () => {
+			const consoleErrorSpy = vi
+				.spyOn(console, 'error')
+				.mockImplementation(() => {});
+			const command = new ListTasksCommand();
+
+			expect((command as any).validateOptions({ format: 'invalid' })).toBe(
+				false
+			);
+			expect(consoleErrorSpy).toHaveBeenCalledWith(
+				expect.stringContaining('Invalid format: invalid')
+			);
+
+			consoleErrorSpy.mockRestore();
+		});
+	});
+
+	describe('--ready filter', () => {
+		it('should filter to only tasks with all dependencies satisfied', async () => {
+			const command = new ListTasksCommand();
+
+			// Mock tasks where some have satisfied deps and some don't
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'done', dependencies: [] },
+				{ id: '2', title: 'Task 2', status: 'pending', dependencies: ['1'] }, // deps satisfied (1 is done)
+				{ id: '3', title: 'Task 3', status: 'pending', dependencies: ['2'] }, // deps NOT satisfied (2 is pending)
+				{ id: '4', title: 'Task 4', status: 'pending', dependencies: [] } // no deps, ready
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 4,
+						filtered: 4,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should only include tasks 2 and 4 (ready to work on)
+			expect(parsed.tasks).toHaveLength(2);
+			expect(parsed.tasks.map((t: any) => t.id)).toEqual(
+				expect.arrayContaining(['2', '4'])
+			);
+			expect(parsed.tasks.map((t: any) => t.id)).not.toContain('3');
+		});
+
+		it('should exclude done/cancelled tasks from ready filter', async () => {
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'done', dependencies: [] },
+				{ id: '2', title: 'Task 2', status: 'cancelled', dependencies: [] },
+				{ id: '3', title: 'Task 3', status: 'pending', dependencies: [] }
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 3,
+						filtered: 3,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should only include task 3 (pending with no deps)
+			expect(parsed.tasks).toHaveLength(1);
+			expect(parsed.tasks[0].id).toBe('3');
+		});
+
+		it('should exclude deferred and blocked tasks from ready filter', async () => {
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'pending', dependencies: [] },
+				{ id: '2', title: 'Task 2', status: 'deferred', dependencies: [] },
+				{ id: '3', title: 'Task 3', status: 'blocked', dependencies: [] },
+				{ id: '4', title: 'Task 4', status: 'in-progress', dependencies: [] },
+				{ id: '5', title: 'Task 5', status: 'review', dependencies: [] }
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 5,
+						filtered: 5,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should only include pending, in-progress, and review tasks
+			expect(parsed.tasks).toHaveLength(3);
+			const ids = parsed.tasks.map((t: any) => t.id);
+			expect(ids).toContain('1'); // pending
+			expect(ids).toContain('4'); // in-progress
+			expect(ids).toContain('5'); // review
+			expect(ids).not.toContain('2'); // deferred - excluded
+			expect(ids).not.toContain('3'); // blocked - excluded
+		});
+	});
+
+	describe('--blocking filter', () => {
+		it('should filter to only tasks that block other tasks', async () => {
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'pending', dependencies: [] }, // blocks 2, 3
+				{ id: '2', title: 'Task 2', status: 'pending', dependencies: ['1'] }, // blocks 4
+				{ id: '3', title: 'Task 3', status: 'pending', dependencies: ['1'] }, // blocks nothing
+				{ id: '4', title: 'Task 4', status: 'pending', dependencies: ['2'] } // blocks nothing
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 4,
+						filtered: 4,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				blocking: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should only include tasks 1 and 2 (they block other tasks)
+			expect(parsed.tasks).toHaveLength(2);
+			expect(parsed.tasks.map((t: any) => t.id)).toEqual(
+				expect.arrayContaining(['1', '2'])
+			);
+		});
+	});
+
+	describe('--ready --blocking combined filter', () => {
+		it('should show high-impact tasks (ready AND blocking)', async () => {
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'done', dependencies: [] },
+				{ id: '2', title: 'Task 2', status: 'pending', dependencies: ['1'] }, // ready (1 done), blocks 3,4
+				{ id: '3', title: 'Task 3', status: 'pending', dependencies: ['2'] }, // not ready, blocks 5
+				{ id: '4', title: 'Task 4', status: 'pending', dependencies: ['2'] }, // not ready, blocks nothing
+				{ id: '5', title: 'Task 5', status: 'pending', dependencies: ['3'] }, // not ready, blocks nothing
+				{ id: '6', title: 'Task 6', status: 'pending', dependencies: [] } // ready, blocks nothing
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 6,
+						filtered: 6,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				blocking: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should only include task 2 (ready AND blocks other tasks)
+			expect(parsed.tasks).toHaveLength(1);
+			expect(parsed.tasks[0].id).toBe('2');
+		});
+	});
+
+	describe('blocks field in output', () => {
+		it('should include blocks field showing which tasks depend on each task', async () => {
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{ id: '1', title: 'Task 1', status: 'pending', dependencies: [] },
+				{ id: '2', title: 'Task 2', status: 'pending', dependencies: ['1'] },
+				{ id: '3', title: 'Task 3', status: 'pending', dependencies: ['1'] },
+				{
+					id: '4',
+					title: 'Task 4',
+					status: 'pending',
+					dependencies: ['2', '3']
+				}
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 4,
+						filtered: 4,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Task 1 blocks tasks 2 and 3
+			const task1 = parsed.tasks.find((t: any) => t.id === '1');
+			expect(task1.blocks).toEqual(expect.arrayContaining(['2', '3']));
+
+			// Task 2 blocks task 4
+			const task2 = parsed.tasks.find((t: any) => t.id === '2');
+			expect(task2.blocks).toEqual(['4']);
+
+			// Task 3 blocks task 4
+			const task3 = parsed.tasks.find((t: any) => t.id === '3');
+			expect(task3.blocks).toEqual(['4']);
+
+			// Task 4 blocks nothing
+			const task4 = parsed.tasks.find((t: any) => t.id === '4');
+			expect(task4.blocks).toEqual([]);
+		});
+	});
+
+	describe('--ready filter edge cases', () => {
+		it('should treat cancelled dependencies as satisfied for --ready filter', async () => {
+			const command = new ListTasksCommand();
+
+			// Task 1 is cancelled, Task 2 depends on Task 1
+			// Task 2 should be considered "ready" because cancelled = complete
+			const mockTasks = [
+				{
+					id: '1',
+					title: 'Cancelled Task',
+					status: 'cancelled',
+					dependencies: []
+				},
+				{
+					id: '2',
+					title: 'Dependent Task',
+					status: 'pending',
+					dependencies: ['1']
+				}
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 2,
+						filtered: 2,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Task 2 should be ready because task 1 (cancelled) counts as complete
+			expect(parsed.tasks).toHaveLength(1);
+			expect(parsed.tasks[0].id).toBe('2');
+			expect(parsed.tasks[0].status).toBe('pending');
+		});
+
+		it('should apply status filter after ready filter', async () => {
+			const command = new ListTasksCommand();
+
+			// Multiple ready tasks with different statuses
+			const mockTasks = [
+				{ id: '1', title: 'Done', status: 'done', dependencies: [] },
+				{
+					id: '2',
+					title: 'Pending ready',
+					status: 'pending',
+					dependencies: ['1']
+				},
+				{
+					id: '3',
+					title: 'In-progress ready',
+					status: 'in-progress',
+					dependencies: ['1']
+				}
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 3,
+						filtered: 3,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				ready: true,
+				status: 'pending',
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// With --ready --status=pending, should only show task 2
+			// Task 2 is ready (dep 1 is done) and pending
+			// Task 3 is ready but in-progress, not pending
+			expect(parsed.tasks).toHaveLength(1);
+			expect(parsed.tasks[0].id).toBe('2');
+			expect(parsed.tasks[0].status).toBe('pending');
+		});
+	});
+
+	describe('buildBlocksMap validation', () => {
+		it('should warn about dependencies to non-existent tasks', async () => {
+			const consoleWarnSpy = vi
+				.spyOn(console, 'warn')
+				.mockImplementation(() => {});
+			const command = new ListTasksCommand();
+
+			const mockTasks = [
+				{
+					id: '1',
+					title: 'Task with bad dep',
+					status: 'pending',
+					dependencies: ['999']
+				}
+			];
+
+			(command as any).tmCore = {
+				tasks: {
+					list: vi.fn().mockResolvedValue({
+						tasks: mockTasks,
+						total: 1,
+						filtered: 1,
+						storageType: 'json'
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				json: true
+			});
+
+			// Verify console.warn was called with warning about invalid dependency
+			expect(consoleWarnSpy).toHaveBeenCalled();
+
+			// Find the call that mentions invalid dependency references
+			const warnCalls = consoleWarnSpy.mock.calls.map((call) => call[0]);
+			const hasInvalidDepWarning = warnCalls.some(
+				(msg) =>
+					typeof msg === 'string' &&
+					msg.includes('invalid dependency reference')
+			);
+			const hasSpecificTaskWarning = warnCalls.some(
+				(msg) =>
+					typeof msg === 'string' &&
+					msg.includes('Task 1') &&
+					msg.includes('999')
+			);
+
+			expect(hasInvalidDepWarning).toBe(true);
+			expect(hasSpecificTaskWarning).toBe(true);
+
+			consoleWarnSpy.mockRestore();
+		});
+	});
+
+	describe('--all-tags option', () => {
+		it('should fetch tasks from multiple tags and include tagName field', async () => {
+			const command = new ListTasksCommand();
+
+			// Mock tasks for different tags
+			const featureATasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Feature A Task 1',
+						status: 'pending',
+						dependencies: []
+					},
+					{
+						id: '2',
+						title: 'Feature A Task 2',
+						status: 'done',
+						dependencies: []
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			const featureBTasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Feature B Task 1',
+						status: 'in-progress',
+						dependencies: []
+					},
+					{
+						id: '2',
+						title: 'Feature B Task 2',
+						status: 'pending',
+						dependencies: ['1']
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			const listMock = vi
+				.fn()
+				.mockResolvedValueOnce(featureATasksResponse)
+				.mockResolvedValueOnce(featureBTasksResponse);
+
+			(command as any).tmCore = {
+				tasks: {
+					list: listMock,
+					getTagsWithStats: vi.fn().mockResolvedValue({
+						tags: [
+							{ name: 'feature-a', taskCount: 2 },
+							{ name: 'feature-b', taskCount: 2 }
+						]
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				allTags: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Should include tasks from both tags
+			expect(parsed.tasks).toHaveLength(4);
+
+			// Each task should have tagName field
+			const featureATasks = parsed.tasks.filter(
+				(t: any) => t.tagName === 'feature-a'
+			);
+			const featureBTasks = parsed.tasks.filter(
+				(t: any) => t.tagName === 'feature-b'
+			);
+
+			expect(featureATasks).toHaveLength(2);
+			expect(featureBTasks).toHaveLength(2);
+
+			// Verify metadata indicates all tags
+			expect(parsed.metadata.allTags).toBe(true);
+			expect(parsed.metadata.tag).toBe('all');
+			expect(parsed.metadata.total).toBe(4);
+		});
+
+		it('should apply --ready filter per-tag when combined with --all-tags', async () => {
+			const command = new ListTasksCommand();
+
+			// Tag A: Task 1 is done, Task 2 depends on Task 1 (ready)
+			const tagATasksResponse = {
+				tasks: [
+					{ id: '1', title: 'Tag A Task 1', status: 'done', dependencies: [] },
+					{
+						id: '2',
+						title: 'Tag A Task 2',
+						status: 'pending',
+						dependencies: ['1']
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			// Tag B: Task 1 is pending (ready), Task 2 depends on Task 1 (not ready)
+			// Note: Task IDs can overlap between tags, but dependencies are tag-scoped
+			const tagBTasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Tag B Task 1',
+						status: 'pending',
+						dependencies: []
+					},
+					{
+						id: '2',
+						title: 'Tag B Task 2',
+						status: 'pending',
+						dependencies: ['1']
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			const listMock = vi
+				.fn()
+				.mockResolvedValueOnce(tagATasksResponse)
+				.mockResolvedValueOnce(tagBTasksResponse);
+
+			(command as any).tmCore = {
+				tasks: {
+					list: listMock,
+					getTagsWithStats: vi.fn().mockResolvedValue({
+						tags: [
+							{ name: 'tag-a', taskCount: 2 },
+							{ name: 'tag-b', taskCount: 2 }
+						]
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				allTags: true,
+				ready: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Tag A: Task 2 is ready (Task 1 is done)
+			// Tag B: Task 1 is ready (no deps), Task 2 is NOT ready (Task 1 is pending)
+			expect(parsed.tasks).toHaveLength(2);
+
+			const taskIds = parsed.tasks.map((t: any) => `${t.tagName}:${t.id}`);
+			expect(taskIds).toContain('tag-a:2'); // Ready: deps satisfied
+			expect(taskIds).toContain('tag-b:1'); // Ready: no deps
+			expect(taskIds).not.toContain('tag-b:2'); // Not ready: depends on pending task
+		});
+
+		it('should reject --all-tags combined with --watch', () => {
+			const consoleErrorSpy = vi
+				.spyOn(console, 'error')
+				.mockImplementation(() => {});
+			const command = new ListTasksCommand();
+
+			const isValid = (command as any).validateOptions({
+				allTags: true,
+				watch: true
+			});
+
+			expect(isValid).toBe(false);
+			expect(consoleErrorSpy).toHaveBeenCalledWith(
+				expect.stringContaining('--all-tags cannot be used with --watch mode')
+			);
+
+			consoleErrorSpy.mockRestore();
+		});
+
+		it('should apply --blocking filter with --all-tags', async () => {
+			const command = new ListTasksCommand();
+
+			// Tag A: Task 1 blocks Task 2
+			const tagATasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Tag A Task 1',
+						status: 'pending',
+						dependencies: []
+					},
+					{
+						id: '2',
+						title: 'Tag A Task 2',
+						status: 'pending',
+						dependencies: ['1']
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			// Tag B: Task 1 blocks nothing (no other tasks depend on it)
+			const tagBTasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Tag B Task 1',
+						status: 'pending',
+						dependencies: []
+					}
+				],
+				total: 1,
+				filtered: 1,
+				storageType: 'json'
+			};
+
+			const listMock = vi
+				.fn()
+				.mockResolvedValueOnce(tagATasksResponse)
+				.mockResolvedValueOnce(tagBTasksResponse);
+
+			(command as any).tmCore = {
+				tasks: {
+					list: listMock,
+					getTagsWithStats: vi.fn().mockResolvedValue({
+						tags: [
+							{ name: 'tag-a', taskCount: 2 },
+							{ name: 'tag-b', taskCount: 1 }
+						]
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				allTags: true,
+				blocking: true,
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Only Tag A Task 1 blocks other tasks
+			expect(parsed.tasks).toHaveLength(1);
+			expect(parsed.tasks[0].tagName).toBe('tag-a');
+			expect(parsed.tasks[0].id).toBe('1');
+			expect(parsed.tasks[0].blocks).toContain('2');
+		});
+
+		it('should apply status filter with --all-tags', async () => {
+			const command = new ListTasksCommand();
+
+			const tagATasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Tag A Task 1',
+						status: 'pending',
+						dependencies: []
+					},
+					{ id: '2', title: 'Tag A Task 2', status: 'done', dependencies: [] }
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			const tagBTasksResponse = {
+				tasks: [
+					{
+						id: '1',
+						title: 'Tag B Task 1',
+						status: 'in-progress',
+						dependencies: []
+					},
+					{
+						id: '2',
+						title: 'Tag B Task 2',
+						status: 'pending',
+						dependencies: []
+					}
+				],
+				total: 2,
+				filtered: 2,
+				storageType: 'json'
+			};
+
+			const listMock = vi
+				.fn()
+				.mockResolvedValueOnce(tagATasksResponse)
+				.mockResolvedValueOnce(tagBTasksResponse);
+
+			(command as any).tmCore = {
+				tasks: {
+					list: listMock,
+					getTagsWithStats: vi.fn().mockResolvedValue({
+						tags: [
+							{ name: 'tag-a', taskCount: 2 },
+							{ name: 'tag-b', taskCount: 2 }
+						]
+					}),
+					getStorageType: vi.fn().mockReturnValue('json')
+				},
+				config: {
+					getActiveTag: vi.fn().mockReturnValue('master')
+				}
+			};
+
+			await (command as any).executeCommand({
+				allTags: true,
+				status: 'pending',
+				json: true
+			});
+
+			const output = consoleLogSpy.mock.calls[0][0];
+			const parsed = JSON.parse(output);
+
+			// Only pending tasks should be included
+			expect(parsed.tasks).toHaveLength(2);
+			expect(parsed.tasks.every((t: any) => t.status === 'pending')).toBe(true);
+
+			const taskIds = parsed.tasks.map((t: any) => `${t.tagName}:${t.id}`);
+			expect(taskIds).toContain('tag-a:1');
+			expect(taskIds).toContain('tag-b:2');
+		});
+	});
+});

apps/cli/src/commands/list.command.ts

@@ -5,14 +5,20 @@
 
 import {
 	OUTPUT_FORMATS,
+	buildBlocksMap,
+	createTmCore,
+	filterBlockingTasks,
+	filterReadyTasks,
+	isTaskComplete,
+	type InvalidDependency,
 	type OutputFormat,
 	STATUS_ICONS,
 	TASK_STATUSES,
 	type Task,
 	type TaskStatus,
+	type TaskWithBlocks,
 	type TmCore,
-	type WatchSubscription,
-	createTmCore
+	type WatchSubscription
 } from '@tm/core';
 import type { StorageType } from '@tm/core';
 import chalk from 'chalk';
@@ -33,7 +39,6 @@ import {
 import { displayCommandHeader } from '../utils/display-helpers.js';
 import { displayError } from '../utils/error-handler.js';
 import { getProjectRoot } from '../utils/project-root.js';
-import { isTaskComplete } from '../utils/task-status.js';
 import * as ui from '../utils/ui.js';
 
 /**
@@ -50,17 +55,26 @@ export interface ListCommandOptions {
 	silent?: boolean;
 	project?: string;
 	watch?: boolean;
+	ready?: boolean;
+	blocking?: boolean;
+	allTags?: boolean;
 }
 
+/**
+ * Task with tag info for cross-tag listing
+ */
+export type TaskWithTag = TaskWithBlocks & { tagName: string };
+
 /**
  * Result type from list command
  */
 export interface ListTasksResult {
-	tasks: Task[];
+	tasks: TaskWithBlocks[] | TaskWithTag[];
 	total: number;
 	filtered: number;
 	tag?: string;
 	storageType: Exclude<StorageType, 'auto'>;
+	allTags?: boolean;
 }
 
 /**
@@ -104,6 +118,15 @@ export class ListTasksCommand extends Command {
 				'Project root directory (auto-detected if not provided)'
 			)
 			.option('-w, --watch', 'Watch for changes and update list automatically')
+			.option(
+				'--ready',
+				'Show only tasks ready to work on (dependencies satisfied)'
+			)
+			.option('--blocking', 'Show only tasks that block other tasks')
+			.option(
+				'--all-tags',
+				'Show tasks from all tags (combine with --ready for actionable tasks)'
+			)
 			.action(async (statusArg?: string, options?: ListCommandOptions) => {
 				// Handle special "all" keyword to show with subtasks
 				let status = statusArg || options?.status;
@@ -142,7 +165,10 @@ export class ListTasksCommand extends Command {
 			if (options.watch) {
 				await this.watchTasks(options);
 			} else {
-				const result = await this.getTasks(options);
+				// Use cross-tag listing when --all-tags is specified
+				const result = options.allTags
+					? await this.getTasksFromAllTags(options)
+					: await this.getTasks(options);
 
 				// Store result for programmatic access
 				this.setLastResult(result);
@@ -195,8 +221,15 @@ export class ListTasksCommand extends Command {
 							// Show sync message with timestamp
 							displaySyncMessage(storageType, lastSync);
 							displayWatchFooter(storageType, lastSync);
-						} catch {
-							// Ignore errors during watch (e.g. partial writes)
+						} catch (refreshError: unknown) {
+							// Log warning but continue watching - don't crash on transient errors
+							const message =
+								refreshError instanceof Error
+									? refreshError.message
+									: String(refreshError);
+							console.error(
+								chalk.yellow(`\nWarning: Failed to refresh tasks: ${message}`)
+							);
 						}
 					} else if (event.type === 'error' && event.error) {
 						console.error(chalk.red(`\n⚠ Watch error: ${event.error.message}`));
@@ -250,6 +283,17 @@ export class ListTasksCommand extends Command {
 			}
 		}
 
+		// Validate --all-tags cannot be used with --watch
+		if (options.allTags && options.watch) {
+			console.error(chalk.red('--all-tags cannot be used with --watch mode'));
+			console.error(
+				chalk.gray(
+					'Use --all-tags without --watch, or --watch without --all-tags'
+				)
+			);
+			return false;
+		}
+
 		return true;
 	}
 
@@ -272,14 +316,20 @@ export class ListTasksCommand extends Command {
 			throw new Error('TmCore not initialized');
 		}
 
-		// Build filter
-		const filter =
+		// Parse status filter values
+		const statusFilterValues =
 			options.status && options.status !== 'all'
-				? {
-						status: options.status
-							.split(',')
-							.map((s: string) => s.trim() as TaskStatus)
-					}
+				? options.status.split(',').map((s: string) => s.trim() as TaskStatus)
+				: undefined;
+
+		// When --ready is used, we need ALL tasks to correctly compute which dependencies are satisfied
+		// So we fetch without status filter first, then apply status filter after ready/blocking
+		const needsAllTasks = options.ready || options.blocking;
+
+		// Build filter - skip status filter if we need all tasks for ready/blocking computation
+		const filter =
+			statusFilterValues && !needsAllTasks
+				? { status: statusFilterValues }
 				: undefined;
 
 		// Call tm-core
@@ -289,7 +339,174 @@ export class ListTasksCommand extends Command {
 			includeSubtasks: options.withSubtasks
 		});
 
-		return result as ListTasksResult;
+		// Build blocks map and enrich tasks with blocks field
+		const { blocksMap, invalidDependencies } = buildBlocksMap(result.tasks);
+		this.displayInvalidDependencyWarnings(invalidDependencies);
+		const enrichedTasks = result.tasks.map((task) => ({
+			...task,
+			blocks: blocksMap.get(String(task.id)) || []
+		}));
+
+		// Apply ready/blocking filters (with full task context)
+		let filteredTasks = enrichedTasks;
+
+		if (options.ready) {
+			filteredTasks = filterReadyTasks(filteredTasks);
+		}
+
+		if (options.blocking) {
+			filteredTasks = filterBlockingTasks(filteredTasks);
+		}
+
+		// Apply status filter AFTER ready/blocking if we deferred it earlier
+		if (statusFilterValues && needsAllTasks) {
+			filteredTasks = filteredTasks.filter((task) =>
+				statusFilterValues.includes(task.status)
+			);
+		}
+
+		return {
+			...result,
+			tasks: filteredTasks,
+			filtered: filteredTasks.length
+		} as ListTasksResult;
+	}
+
+	/**
+	 * Get ready tasks from all tags
+	 * Fetches tasks from each tag and combines them with tag info
+	 */
+	private async getTasksFromAllTags(
+		options: ListCommandOptions
+	): Promise<ListTasksResult> {
+		if (!this.tmCore) {
+			throw new Error('TmCore not initialized');
+		}
+
+		// Get all tags
+		const tagsResult = await this.tmCore.tasks.getTagsWithStats();
+		const allTaggedTasks: TaskWithTag[] = [];
+		let totalTaskCount = 0;
+		const failedTags: Array<{ name: string; error: string }> = [];
+
+		// Fetch tasks from each tag
+		for (const tagInfo of tagsResult.tags) {
+			const tagName = tagInfo.name;
+
+			try {
+				// Get tasks for this tag
+				const result = await this.tmCore.tasks.list({
+					tag: tagName,
+					includeSubtasks: options.withSubtasks
+				});
+
+				// Track total count before any filtering (consistent with getTasks)
+				totalTaskCount += result.tasks.length;
+
+				// Build blocks map for this tag's tasks
+				const { blocksMap, invalidDependencies } = buildBlocksMap(result.tasks);
+				this.displayInvalidDependencyWarnings(invalidDependencies, tagName);
+
+				// Enrich tasks with blocks field and tag name
+				const enrichedTasks: TaskWithTag[] = result.tasks.map((task) => ({
+					...task,
+					blocks: blocksMap.get(String(task.id)) || [],
+					tagName
+				}));
+
+				// Apply ready filter per-tag to respect tag-scoped dependencies
+				// (task IDs may overlap between tags, so we must filter within each tag)
+				const tasksToAdd: TaskWithTag[] = options.ready
+					? (filterReadyTasks(enrichedTasks) as TaskWithTag[])
+					: enrichedTasks;
+
+				allTaggedTasks.push(...tasksToAdd);
+			} catch (tagError: unknown) {
+				const errorMessage =
+					tagError instanceof Error ? tagError.message : String(tagError);
+				failedTags.push({ name: tagName, error: errorMessage });
+				continue; // Skip this tag but continue with others
+			}
+		}
+
+		// Warn about failed tags
+		if (failedTags.length > 0) {
+			console.warn(
+				chalk.yellow(
+					`\nWarning: Could not fetch tasks from ${failedTags.length} tag(s):`
+				)
+			);
+			failedTags.forEach(({ name, error }) => {
+				console.warn(chalk.gray(`  ${name}: ${error}`));
+			});
+		}
+
+		// If ALL tags failed, throw to surface the issue
+		if (
+			failedTags.length === tagsResult.tags.length &&
+			tagsResult.tags.length > 0
+		) {
+			throw new Error(
+				`Failed to fetch tasks from any tag. First error: ${failedTags[0].error}`
+			);
+		}
+
+		// Apply additional filters
+		let filteredTasks: TaskWithTag[] = allTaggedTasks;
+
+		// Apply blocking filter if specified
+		if (options.blocking) {
+			filteredTasks = filteredTasks.filter((task) => task.blocks.length > 0);
+		}
+
+		// Apply status filter if specified
+		if (options.status && options.status !== 'all') {
+			const statusValues = options.status
+				.split(',')
+				.map((s) => s.trim() as TaskStatus);
+			filteredTasks = filteredTasks.filter((task) =>
+				statusValues.includes(task.status)
+			);
+		}
+
+		return {
+			tasks: filteredTasks,
+			total: totalTaskCount,
+			filtered: filteredTasks.length,
+			storageType: this.tmCore.tasks.getStorageType(),
+			allTags: true
+		};
+	}
+
+	/**
+	 * Display warnings for invalid dependency references
+	 * @param invalidDependencies - Array of invalid dependency references from buildBlocksMap
+	 * @param tagName - Optional tag name for context in multi-tag mode
+	 */
+	private displayInvalidDependencyWarnings(
+		invalidDependencies: InvalidDependency[],
+		tagName?: string
+	): void {
+		if (invalidDependencies.length === 0) {
+			return;
+		}
+
+		const tagContext = tagName ? ` (tag: ${tagName})` : '';
+		console.warn(
+			chalk.yellow(
+				`\nWarning: ${invalidDependencies.length} invalid dependency reference(s) found${tagContext}:`
+			)
+		);
+		invalidDependencies.slice(0, 5).forEach(({ taskId, depId }) => {
+			console.warn(
+				chalk.gray(`  Task ${taskId} depends on non-existent task ${depId}`)
+			);
+		});
+		if (invalidDependencies.length > 5) {
+			console.warn(
+				chalk.gray(`  ...and ${invalidDependencies.length - 5} more`)
+			);
+		}
 	}
 
 	/**
@@ -300,13 +517,12 @@ export class ListTasksCommand extends Command {
 		options: ListCommandOptions
 	): void {
 		// Resolve format: --json and --compact flags override --format option
-		const format = (
-			options.json
-				? 'json'
-				: options.compact
-					? 'compact'
-					: options.format || 'text'
-		) as OutputFormat;
+		let format: OutputFormat = options.format || 'text';
+		if (options.json) {
+			format = 'json';
+		} else if (options.compact) {
+			format = 'compact';
+		}
 
 		switch (format) {
 			case 'json':
@@ -335,8 +551,9 @@ export class ListTasksCommand extends Command {
 					metadata: {
 						total: data.total,
 						filtered: data.filtered,
-						tag: data.tag,
-						storageType: data.storageType
+						tag: data.allTags ? 'all' : data.tag,
+						storageType: data.storageType,
+						allTags: data.allTags || false
 					}
 				},
 				null,
@@ -352,29 +569,32 @@ export class ListTasksCommand extends Command {
 		data: ListTasksResult,
 		options: ListCommandOptions
 	): void {
-		const { tasks, tag, storageType } = data;
+		const { tasks, tag, storageType, allTags } = data;
 
 		// Display header unless --no-header is set
 		if (options.noHeader !== true) {
 			displayCommandHeader(this.tmCore, {
-				tag: tag || 'master',
+				tag: allTags ? 'all tags' : tag || 'master',
 				storageType
 			});
 		}
 
-		tasks.forEach((task) => {
+		for (const task of tasks) {
 			const icon = STATUS_ICONS[task.status];
-			console.log(`${chalk.cyan(task.id)} ${icon} ${task.title}`);
+			// Show tag in compact format when listing all tags (tasks are TaskWithTag[])
+			const tagPrefix =
+				allTags && 'tagName' in task ? chalk.magenta(`[${task.tagName}] `) : '';
+			console.log(`${tagPrefix}${chalk.cyan(task.id)} ${icon} ${task.title}`);
 
 			if (options.withSubtasks && task.subtasks?.length) {
-				task.subtasks.forEach((subtask) => {
+				for (const subtask of task.subtasks) {
 					const subIcon = STATUS_ICONS[subtask.status];
 					console.log(
 						`  ${chalk.gray(String(subtask.id))} ${subIcon} ${chalk.gray(subtask.title)}`
 					);
-				});
+				}
 			}
-		});
+		}
 	}
 
 	/**
@@ -384,12 +604,12 @@ export class ListTasksCommand extends Command {
 		data: ListTasksResult,
 		options: ListCommandOptions
 	): void {
-		const { tasks, tag, storageType } = data;
+		const { tasks, tag, storageType, allTags } = data;
 
 		// Display header unless --no-header is set
 		if (options.noHeader !== true) {
 			displayCommandHeader(this.tmCore, {
-				tag: tag || 'master',
+				tag: allTags ? 'all tags' : tag || 'master',
 				storageType
 			});
 		}
@@ -414,43 +634,52 @@ export class ListTasksCommand extends Command {
 			? tasks.find((t) => String(t.id) === String(nextTaskInfo.id))
 			: undefined;
 
-		// Display dashboard boxes (nextTask already has complexity from storage enrichment)
-		displayDashboards(
-			taskStats,
-			subtaskStats,
-			priorityBreakdown,
-			depStats,
-			nextTask
-		);
+		// Display dashboard boxes unless filtering by --ready, --blocking, or --all-tags
+		// (filtered/cross-tag dashboards would show misleading statistics)
+		const isFiltered = options.ready || options.blocking || allTags;
+		if (!isFiltered) {
+			displayDashboards(
+				taskStats,
+				subtaskStats,
+				priorityBreakdown,
+				depStats,
+				nextTask
+			);
+		}
 
 		// Task table
 		console.log(
 			ui.createTaskTable(tasks, {
 				showSubtasks: options.withSubtasks,
 				showDependencies: true,
-				showComplexity: true // Enable complexity column
+				showBlocks: true, // Show which tasks this one blocks
+				showComplexity: true, // Enable complexity column
+				showTag: allTags // Show tag column for cross-tag listing
 			})
 		);
 
 		// Display recommended next task section immediately after table
-		// Don't show "no tasks available" message in list command - that's for tm next
-		if (nextTask) {
-			const description = getTaskDescription(nextTask);
+		// Skip when filtering by --ready or --blocking (user already knows what they're looking at)
+		if (!isFiltered) {
+			// Don't show "no tasks available" message in list command - that's for tm next
+			if (nextTask) {
+				const description = getTaskDescription(nextTask);
 
-			displayRecommendedNextTask({
-				id: nextTask.id,
-				title: nextTask.title,
-				priority: nextTask.priority,
-				status: nextTask.status,
-				dependencies: nextTask.dependencies,
-				description,
-				complexity: nextTask.complexity as number | undefined
-			});
+				displayRecommendedNextTask({
+					id: nextTask.id,
+					title: nextTask.title,
+					priority: nextTask.priority,
+					status: nextTask.status,
+					dependencies: nextTask.dependencies,
+					description,
+					complexity: nextTask.complexity as number | undefined
+				});
+			}
+			// If no next task, don't show any message - dashboard already shows the info
+
+			// Display suggested next steps at the end
+			displaySuggestedNextSteps();
 		}
-		// If no next task, don't show any message - dashboard already shows the info
-
-		// Display suggested next steps at the end
-		displaySuggestedNextSteps();
 	}
 
 	/**

apps/cli/src/commands/loop.command.spec.ts

@@ -77,8 +77,8 @@ describe('LoopCommand', () => {
 		mockTmCore = {
 			loop: {
 				run: mockLoopRun,
-				checkSandboxAuth: vi.fn().mockReturnValue(true),
-				runInteractiveAuth: vi.fn(),
+				checkSandboxAuth: vi.fn().mockReturnValue({ ready: true }),
+				runInteractiveAuth: vi.fn().mockReturnValue({ success: true }),
 				resolveIterations: vi.fn().mockImplementation((opts) => {
 					// Mirror the real implementation logic for accurate testing
 					if (opts.userIterations !== undefined) return opts.userIterations;
@@ -389,27 +389,79 @@ describe('LoopCommand', () => {
 			);
 		});
 
-		it('should check sandbox auth before running', async () => {
+		it('should check sandbox auth when --sandbox flag is provided', async () => {
 			const result = createMockResult();
 			mockLoopRun.mockResolvedValue(result);
+			mockTmCore.loop.checkSandboxAuth.mockReturnValue({ ready: true });
 
 			const execute = (loopCommand as any).execute.bind(loopCommand);
-			await execute({});
+			await execute({ sandbox: true });
 
 			expect(mockTmCore.loop.checkSandboxAuth).toHaveBeenCalled();
 		});
 
-		it('should run interactive auth when sandbox not ready', async () => {
-			mockTmCore.loop.checkSandboxAuth.mockReturnValue(false);
+		it('should not check sandbox auth when --sandbox flag is not provided', async () => {
 			const result = createMockResult();
 			mockLoopRun.mockResolvedValue(result);
 
 			const execute = (loopCommand as any).execute.bind(loopCommand);
 			await execute({});
 
+			expect(mockTmCore.loop.checkSandboxAuth).not.toHaveBeenCalled();
+		});
+
+		it('should run interactive auth when sandbox not ready', async () => {
+			mockTmCore.loop.checkSandboxAuth.mockReturnValue({ ready: false });
+			mockTmCore.loop.runInteractiveAuth.mockReturnValue({ success: true });
+			const result = createMockResult();
+			mockLoopRun.mockResolvedValue(result);
+
+			const execute = (loopCommand as any).execute.bind(loopCommand);
+			await execute({ sandbox: true });
+
 			expect(mockTmCore.loop.runInteractiveAuth).toHaveBeenCalled();
 		});
 
+		it('should throw error when sandbox auth has error', async () => {
+			mockTmCore.loop.checkSandboxAuth.mockReturnValue({
+				error: 'Sandbox auth failed'
+			});
+
+			const execute = (loopCommand as any).execute.bind(loopCommand);
+
+			try {
+				await execute({ sandbox: true });
+			} catch {
+				// Expected - processExitSpy mock throws to simulate process.exit
+			}
+
+			expect(displayError).toHaveBeenCalledWith(
+				expect.objectContaining({ message: 'Sandbox auth failed' }),
+				{ skipExit: true }
+			);
+		});
+
+		it('should throw error when interactive auth fails', async () => {
+			mockTmCore.loop.checkSandboxAuth.mockReturnValue({ ready: false });
+			mockTmCore.loop.runInteractiveAuth.mockReturnValue({
+				success: false,
+				error: 'Auth failed'
+			});
+
+			const execute = (loopCommand as any).execute.bind(loopCommand);
+
+			try {
+				await execute({ sandbox: true });
+			} catch {
+				// Expected - processExitSpy mock throws to simulate process.exit
+			}
+
+			expect(displayError).toHaveBeenCalledWith(
+				expect.objectContaining({ message: 'Auth failed' }),
+				{ skipExit: true }
+			);
+		});
+
 		it('should show next task before starting loop', async () => {
 			const result = createMockResult();
 			mockLoopRun.mockResolvedValue(result);

apps/cli/src/commands/loop.command.ts

@@ -6,9 +6,9 @@ import path from 'node:path';
 import {
 	type LoopConfig,
 	type LoopResult,
+	PRESET_NAMES,
 	type TmCore,
-	createTmCore,
-	PRESET_NAMES
+	createTmCore
 } from '@tm/core';
 import chalk from 'chalk';
 import { Command } from 'commander';
@@ -22,6 +22,7 @@ export interface LoopCommandOptions {
 	progressFile?: string;
 	tag?: string;
 	project?: string;
+	sandbox?: boolean;
 }
 
 export class LoopCommand extends Command {
@@ -47,6 +48,7 @@ export class LoopCommand extends Command {
 				'--project <path>',
 				'Project root directory (auto-detected if not provided)'
 			)
+			.option('--sandbox', 'Run Claude in Docker sandbox mode')
 			.action((options: LoopCommandOptions) => this.execute(options));
 	}
 
@@ -80,11 +82,17 @@ export class LoopCommand extends Command {
 				storageType: this.tmCore.tasks.getStorageType()
 			});
 
-			this.handleSandboxAuth();
+			// Only check sandbox auth when --sandbox flag is used
+			if (options.sandbox) {
+				this.handleSandboxAuth();
+			}
 
 			console.log(chalk.cyan('Starting Task Master Loop...'));
 			console.log(chalk.dim(`Preset: ${prompt}`));
 			console.log(chalk.dim(`Max iterations: ${iterations}`));
+			console.log(
+				chalk.dim(`Mode: ${options.sandbox ? 'Docker sandbox' : 'Claude CLI'}`)
+			);
 
 			// Show next task only for default preset (other presets don't use Task Master tasks)
 			if (prompt === 'default') {
@@ -105,7 +113,8 @@ export class LoopCommand extends Command {
 				iterations,
 				prompt,
 				progressFile,
-				tag: options.tag
+				tag: options.tag,
+				sandbox: options.sandbox
 			};
 
 			const result = await this.tmCore.loop.run(config);
@@ -118,9 +127,13 @@ export class LoopCommand extends Command {
 
 	private handleSandboxAuth(): void {
 		console.log(chalk.dim('Checking sandbox auth...'));
-		const isAuthed = this.tmCore.loop.checkSandboxAuth();
+		const authCheck = this.tmCore.loop.checkSandboxAuth();
 
-		if (isAuthed) {
+		if (authCheck.error) {
+			throw new Error(authCheck.error);
+		}
+
+		if (authCheck.ready) {
 			console.log(chalk.green('✓ Sandbox ready'));
 			return;
 		}
@@ -132,7 +145,10 @@ export class LoopCommand extends Command {
 		);
 		console.log(chalk.dim('Please complete auth, then Ctrl+C to continue.\n'));
 
-		this.tmCore.loop.runInteractiveAuth();
+		const authResult = this.tmCore.loop.runInteractiveAuth();
+		if (!authResult.success) {
+			throw new Error(authResult.error || 'Interactive authentication failed');
+		}
 		console.log(chalk.green('✓ Auth complete\n'));
 	}
 

apps/cli/src/commands/tags.command.ts

@@ -77,8 +77,10 @@ export class TagsCommand extends Command {
 	constructor(name?: string) {
 		super(name || 'tags');
 
-		// Configure the command
-		this.description('Manage tags for task organization');
+		// Configure the command with options that apply to default list action
+		this.description('Manage tags for task organization')
+			.option('--show-metadata', 'Show additional tag metadata')
+			.option('--ready', 'Show only tags with ready tasks available');
 
 		// Add subcommands
 		this.addListCommand();
@@ -88,9 +90,9 @@ export class TagsCommand extends Command {
 		this.addRenameCommand();
 		this.addCopyCommand();
 
-		// Default action: list tags
-		this.action(async () => {
-			await this.executeList();
+		// Default action: list tags (with options from parent command)
+		this.action(async (options) => {
+			await this.executeList(options);
 		});
 	}
 
@@ -101,6 +103,7 @@ export class TagsCommand extends Command {
 		this.command('list')
 			.description('List all tags with statistics (default action)')
 			.option('--show-metadata', 'Show additional tag metadata')
+			.option('--ready', 'Show only tags with ready tasks available')
 			.addHelpText(
 				'after',
 				`
@@ -108,6 +111,7 @@ Examples:
   $ tm tags          # List all tags (default)
   $ tm tags list     # List all tags (explicit)
   $ tm tags list --show-metadata  # List with metadata
+  $ tm tags list --ready  # Show only tags with parallelizable work
 `
 			)
 			.action(async (options) => {
@@ -245,6 +249,7 @@ Examples:
 	 */
 	private async executeList(options?: {
 		showMetadata?: boolean;
+		ready?: boolean;
 	}): Promise<void> {
 		try {
 			// Initialize tmCore first (needed by bridge functions)
@@ -257,7 +262,8 @@ Examples:
 				tasksPath,
 				{
 					showTaskCounts: true,
-					showMetadata: options?.showMetadata || false
+					showMetadata: options?.showMetadata || false,
+					ready: options?.ready || false
 				},
 				{ projectRoot },
 				'text'

apps/cli/src/ui/display/tables.ts

@@ -6,6 +6,7 @@
 import type { Subtask, Task, TaskPriority } from '@tm/core';
 import chalk from 'chalk';
 import Table from 'cli-table3';
+import { isTaskComplete } from '../../utils/task-status.js';
 import { getComplexityWithColor } from '../formatters/complexity-formatters.js';
 import { getPriorityWithColor } from '../formatters/priority-formatters.js';
 import { getStatusWithColor } from '../formatters/status-formatters.js';
@@ -16,59 +17,94 @@ import { getBoxWidth, truncate } from '../layout/helpers.js';
  */
 const DEFAULT_PRIORITY: TaskPriority = 'medium';
 
+/**
+ * Task-like object that can optionally have blocks field and tag name
+ * Used for table display - accepts both enriched TaskWithBlocks and regular Task/Subtask
+ */
+export type TaskTableItem = (Task | Subtask) & {
+	blocks?: string[];
+	tagName?: string;
+};
+
+/**
+ * Column width ratios indexed by number of optional columns
+ * Each array contains ratios for: [Tag?, ID, Title, Status, Priority, Dependencies?, Blocks?, Complexity?]
+ */
+const COLUMN_WIDTH_RATIOS: Record<number, number[]> = {
+	0: [0.1, 0.5, 0.2, 0.2],
+	1: [0.08, 0.4, 0.18, 0.14, 0.2],
+	2: [0.08, 0.35, 0.14, 0.11, 0.16, 0.16],
+	3: [0.07, 0.3, 0.12, 0.1, 0.14, 0.14, 0.1],
+	4: [0.12, 0.06, 0.2, 0.1, 0.1, 0.12, 0.12, 0.1]
+};
+
 /**
  * Create a task table for display
  */
 export function createTaskTable(
-	tasks: (Task | Subtask)[],
+	tasks: TaskTableItem[],
 	options?: {
 		showSubtasks?: boolean;
 		showComplexity?: boolean;
 		showDependencies?: boolean;
+		showBlocks?: boolean;
+		showTag?: boolean;
 	}
 ): string {
 	const {
 		showSubtasks = false,
 		showComplexity = false,
-		showDependencies = true
+		showDependencies = true,
+		showBlocks = false,
+		showTag = false
 	} = options || {};
 
 	// Calculate dynamic column widths based on terminal width
 	const tableWidth = getBoxWidth(0.9, 100);
-	// Adjust column widths to better match the original layout
-	const baseColWidths = showComplexity
-		? [
-				Math.floor(tableWidth * 0.1),
-				Math.floor(tableWidth * 0.4),
-				Math.floor(tableWidth * 0.15),
-				Math.floor(tableWidth * 0.1),
-				Math.floor(tableWidth * 0.2),
-				Math.floor(tableWidth * 0.1)
-			] // ID, Title, Status, Priority, Dependencies, Complexity
-		: [
-				Math.floor(tableWidth * 0.08),
-				Math.floor(tableWidth * 0.4),
-				Math.floor(tableWidth * 0.18),
-				Math.floor(tableWidth * 0.12),
-				Math.floor(tableWidth * 0.2)
-			]; // ID, Title, Status, Priority, Dependencies
 
-	const headers = [
+	// Count optional columns and get corresponding width ratios
+	const optionalCols =
+		(showTag ? 1 : 0) +
+		(showDependencies ? 1 : 0) +
+		(showBlocks ? 1 : 0) +
+		(showComplexity ? 1 : 0);
+
+	const ratios = COLUMN_WIDTH_RATIOS[optionalCols] || COLUMN_WIDTH_RATIOS[4];
+	const baseColWidths = ratios.map((ratio) => Math.floor(tableWidth * ratio));
+
+	// Build headers and column widths dynamically
+	const headers: string[] = [];
+	const colWidths: number[] = [];
+	let colIndex = 0;
+
+	if (showTag) {
+		headers.push(chalk.blue.bold('Tag'));
+		colWidths.push(baseColWidths[colIndex++]);
+	}
+
+	// Core columns: ID, Title, Status, Priority
+	headers.push(
 		chalk.blue.bold('ID'),
 		chalk.blue.bold('Title'),
 		chalk.blue.bold('Status'),
 		chalk.blue.bold('Priority')
-	];
-	const colWidths = baseColWidths.slice(0, 4);
+	);
+	colWidths.push(...baseColWidths.slice(colIndex, colIndex + 4));
+	colIndex += 4;
 
 	if (showDependencies) {
 		headers.push(chalk.blue.bold('Dependencies'));
-		colWidths.push(baseColWidths[4]);
+		colWidths.push(baseColWidths[colIndex++]);
+	}
+
+	if (showBlocks) {
+		headers.push(chalk.blue.bold('Blocks'));
+		colWidths.push(baseColWidths[colIndex++]);
 	}
 
 	if (showComplexity) {
 		headers.push(chalk.blue.bold('Complexity'));
-		colWidths.push(baseColWidths[5] || 12);
+		colWidths.push(baseColWidths[colIndex] || 12);
 	}
 
 	const table = new Table({
@@ -79,17 +115,27 @@ export function createTaskTable(
 	});
 
 	tasks.forEach((task) => {
-		const row: string[] = [
+		const row: string[] = [];
+
+		// Tag goes first when showing all tags
+		if (showTag) {
+			row.push(chalk.magenta(task.tagName || '-'));
+		}
+
+		// Core columns: ID, Title, Status, Priority
+		// Title column index depends on whether tag is shown
+		const titleColIndex = showTag ? 2 : 1;
+		row.push(
 			chalk.cyan(task.id.toString()),
-			truncate(task.title, colWidths[1] - 3),
+			truncate(task.title, colWidths[titleColIndex] - 3),
 			getStatusWithColor(task.status, true), // Use table version
 			getPriorityWithColor(task.priority)
-		];
+		);
 
 		if (showDependencies) {
 			// For table display, show simple format without status icons
 			if (!task.dependencies || task.dependencies.length === 0) {
-				row.push(chalk.gray('None'));
+				row.push(chalk.gray('-'));
 			} else {
 				row.push(
 					chalk.cyan(task.dependencies.map((d) => String(d)).join(', '))
@@ -97,6 +143,21 @@ export function createTaskTable(
 			}
 		}
 
+		if (showBlocks) {
+			// Show tasks that depend on this one
+			if (!task.blocks || task.blocks.length === 0) {
+				row.push(chalk.gray('-'));
+			} else {
+				// Gray out blocks for completed tasks (no longer blocking)
+				const blocksText = task.blocks.join(', ');
+				row.push(
+					isTaskComplete(task.status)
+						? chalk.gray(blocksText)
+						: chalk.yellow(blocksText)
+				);
+			}
+		}
+
 		if (showComplexity) {
 			// Show complexity score from report if available
 			if (typeof task.complexity === 'number') {
@@ -111,23 +172,38 @@ export function createTaskTable(
 		// Add subtasks if requested
 		if (showSubtasks && task.subtasks && task.subtasks.length > 0) {
 			task.subtasks.forEach((subtask) => {
-				const subRow: string[] = [
+				const subRow: string[] = [];
+
+				// Tag goes first when showing all tags
+				if (showTag) {
+					// Subtasks inherit parent's tag, just show dash
+					subRow.push(chalk.gray('-'));
+				}
+
+				// Core subtask columns: ID, Title, Status, Priority
+				const subTitleColIndex = showTag ? 2 : 1;
+				subRow.push(
 					chalk.gray(` └─ ${subtask.id}`),
-					chalk.gray(truncate(subtask.title, colWidths[1] - 6)),
+					chalk.gray(truncate(subtask.title, colWidths[subTitleColIndex] - 6)),
 					chalk.gray(getStatusWithColor(subtask.status, true)),
 					chalk.gray(subtask.priority || DEFAULT_PRIORITY)
-				];
+				);
 
 				if (showDependencies) {
 					subRow.push(
 						chalk.gray(
 							subtask.dependencies && subtask.dependencies.length > 0
 								? subtask.dependencies.map((dep) => String(dep)).join(', ')
-								: 'None'
+								: '-'
 						)
 					);
 				}
 
+				if (showBlocks) {
+					// Subtasks don't typically have blocks, show dash
+					subRow.push(chalk.gray('-'));
+				}
+
 				if (showComplexity) {
 					const complexityDisplay =
 						typeof subtask.complexity === 'number'

apps/cli/tests/integration/commands/loop.command.test.ts

@@ -14,9 +14,12 @@ import fs from 'node:fs';
 import os from 'node:os';
 import path from 'node:path';
 import { createTask, createTasksFile } from '@tm/core/testing';
-import { afterEach, beforeEach, describe, expect, it } from 'vitest';
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 import { getCliBinPath } from '../../helpers/test-utils.js';
 
+// Increase hook timeout for this file - init command can be slow in CI
+vi.setConfig({ hookTimeout: 30000 });
+
 // Capture initial working directory at module load time
 const initialCwd = process.cwd();
 
@@ -144,11 +147,11 @@ describe('loop command', () => {
 			expect(output).toContain('--tag');
 		});
 
-		it('should show --json option in help', () => {
+		it('should show --sandbox option in help', () => {
 			const { output, exitCode } = runHelp();
 
 			expect(exitCode).toBe(0);
-			expect(output).toContain('--json');
+			expect(output).toContain('--sandbox');
 		});
 
 		it('should show --progress-file option in help', () => {
@@ -191,44 +194,6 @@ describe('loop command', () => {
 		});
 	});
 
-	describe('option parsing', () => {
-		it('should accept valid iterations', () => {
-			// Command will fail when trying to run claude, but validation should pass
-			const { output } = runLoop('-n 5');
-
-			// Should NOT contain validation error for iterations
-			expect(output.toLowerCase()).not.toContain('invalid iterations');
-		});
-
-		it('should accept custom prompt preset', () => {
-			const { output } = runLoop('-p test-coverage');
-
-			// Should NOT contain validation error for prompt
-			expect(output.toLowerCase()).not.toContain('invalid prompt');
-		});
-
-		it('should accept tag filter', () => {
-			const { output } = runLoop('-t feature');
-
-			// Should NOT contain validation error for tag
-			expect(output.toLowerCase()).not.toContain('invalid tag');
-		});
-
-		it('should accept progress-file option', () => {
-			const { output } = runLoop('--progress-file /tmp/test-progress.txt');
-
-			// Should NOT contain validation error for progress-file
-			expect(output.toLowerCase()).not.toContain('invalid progress');
-		});
-
-		it('should accept multiple options together', () => {
-			const { output } = runLoop('-n 3 -p default -t test');
-
-			// Should NOT contain validation errors
-			expect(output.toLowerCase()).not.toContain('invalid iterations');
-		});
-	});
-
 	describe('error messages', () => {
 		it('should show helpful error for invalid iterations', () => {
 			const { output, exitCode } = runLoop('-n invalid');
@@ -239,23 +204,4 @@ describe('loop command', () => {
 			expect(output.toLowerCase()).toContain('positive');
 		});
 	});
-
-	describe('project detection', () => {
-		it('should work in initialized project directory', () => {
-			// The project is already initialized in beforeEach
-			// Command will fail when trying to run claude, but project detection should work
-			const { output } = runLoop('-n 1');
-
-			// Should NOT contain "not a task-master project" or similar
-			expect(output.toLowerCase()).not.toContain('not initialized');
-			expect(output.toLowerCase()).not.toContain('no project');
-		});
-
-		it('should accept --project option for explicit path', () => {
-			const { output } = runLoop(`--project "${testDir}" -n 1`);
-
-			// Should NOT contain validation error for project path
-			expect(output.toLowerCase()).not.toContain('invalid project');
-		});
-	});
 });

apps/cli/tests/unit/commands/list.command.spec.ts

@@ -1,195 +0,0 @@
-/**
- * @fileoverview Unit tests for ListTasksCommand
- */
-
-import type { TmCore } from '@tm/core';
-import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
-
-// Mock dependencies
-vi.mock('@tm/core', () => ({
-	createTmCore: vi.fn(),
-	OUTPUT_FORMATS: ['text', 'json', 'compact'],
-	TASK_STATUSES: [
-		'pending',
-		'in-progress',
-		'done',
-		'review',
-		'deferred',
-		'cancelled'
-	],
-	STATUS_ICONS: {
-		pending: '⏳',
-		'in-progress': '🔄',
-		done: '✅',
-		review: '👀',
-		deferred: '⏸️',
-		cancelled: '❌'
-	}
-}));
-
-vi.mock('../../../src/utils/project-root.js', () => ({
-	getProjectRoot: vi.fn((path?: string) => path || '/test/project')
-}));
-
-vi.mock('../../../src/utils/error-handler.js', () => ({
-	displayError: vi.fn()
-}));
-
-vi.mock('../../../src/utils/display-helpers.js', () => ({
-	displayCommandHeader: vi.fn()
-}));
-
-vi.mock('../../../src/ui/index.js', () => ({
-	calculateDependencyStatistics: vi.fn(() => ({ total: 0, blocked: 0 })),
-	calculateSubtaskStatistics: vi.fn(() => ({ total: 0, completed: 0 })),
-	calculateTaskStatistics: vi.fn(() => ({ total: 0, completed: 0 })),
-	displayDashboards: vi.fn(),
-	displayRecommendedNextTask: vi.fn(),
-	displaySuggestedNextSteps: vi.fn(),
-	getPriorityBreakdown: vi.fn(() => ({})),
-	getTaskDescription: vi.fn(() => 'Test description')
-}));
-
-vi.mock('../../../src/utils/ui.js', () => ({
-	createTaskTable: vi.fn(() => 'Table output'),
-	displayWarning: vi.fn()
-}));
-
-import { ListTasksCommand } from '../../../src/commands/list.command.js';
-
-describe('ListTasksCommand', () => {
-	let consoleLogSpy: any;
-	let mockTmCore: Partial<TmCore>;
-
-	beforeEach(() => {
-		consoleLogSpy = vi.spyOn(console, 'log').mockImplementation(() => {});
-
-		mockTmCore = {
-			tasks: {
-				list: vi.fn().mockResolvedValue({
-					tasks: [{ id: '1', title: 'Test Task', status: 'pending' }],
-					total: 1,
-					filtered: 1,
-					storageType: 'json'
-				}),
-				getStorageType: vi.fn().mockReturnValue('json')
-			} as any,
-			config: {
-				getActiveTag: vi.fn().mockReturnValue('master')
-			} as any
-		};
-	});
-
-	afterEach(() => {
-		vi.clearAllMocks();
-		consoleLogSpy.mockRestore();
-	});
-
-	describe('JSON output format', () => {
-		it('should use JSON format when --json flag is set', async () => {
-			const command = new ListTasksCommand();
-
-			// Mock the tmCore initialization
-			(command as any).tmCore = mockTmCore;
-
-			// Execute with --json flag
-			await (command as any).executeCommand({
-				json: true,
-				format: 'text' // Should be overridden by --json
-			});
-
-			// Verify JSON output was called
-			expect(consoleLogSpy).toHaveBeenCalled();
-			const output = consoleLogSpy.mock.calls[0][0];
-
-			// Should be valid JSON
-			expect(() => JSON.parse(output)).not.toThrow();
-
-			const parsed = JSON.parse(output);
-			expect(parsed).toHaveProperty('tasks');
-			expect(parsed).toHaveProperty('metadata');
-		});
-
-		it('should override --format when --json is set', async () => {
-			const command = new ListTasksCommand();
-			(command as any).tmCore = mockTmCore;
-
-			await (command as any).executeCommand({
-				json: true,
-				format: 'compact' // Should be overridden
-			});
-
-			// Should output JSON, not compact format
-			const output = consoleLogSpy.mock.calls[0][0];
-			expect(() => JSON.parse(output)).not.toThrow();
-		});
-
-		it('should use specified format when --json is not set', async () => {
-			const command = new ListTasksCommand();
-			(command as any).tmCore = mockTmCore;
-
-			await (command as any).executeCommand({
-				format: 'compact'
-			});
-
-			// Should use compact format (not JSON)
-			const output = consoleLogSpy.mock.calls;
-			// In compact mode, output is not JSON
-			expect(output.length).toBeGreaterThan(0);
-		});
-
-		it('should default to text format when neither flag is set', async () => {
-			const command = new ListTasksCommand();
-			(command as any).tmCore = mockTmCore;
-
-			await (command as any).executeCommand({});
-
-			// Should use text format (not JSON)
-			// If any console.log was called, verify it's not JSON
-			if (consoleLogSpy.mock.calls.length > 0) {
-				const output = consoleLogSpy.mock.calls[0][0];
-				// Text format output should not be parseable JSON
-				// or should be the table string we mocked
-				expect(
-					output === 'Table output' ||
-						(() => {
-							try {
-								JSON.parse(output);
-								return false;
-							} catch {
-								return true;
-							}
-						})()
-				).toBe(true);
-			}
-		});
-	});
-
-	describe('format validation', () => {
-		it('should accept valid formats', () => {
-			const command = new ListTasksCommand();
-
-			expect((command as any).validateOptions({ format: 'text' })).toBe(true);
-			expect((command as any).validateOptions({ format: 'json' })).toBe(true);
-			expect((command as any).validateOptions({ format: 'compact' })).toBe(
-				true
-			);
-		});
-
-		it('should reject invalid formats', () => {
-			const consoleErrorSpy = vi
-				.spyOn(console, 'error')
-				.mockImplementation(() => {});
-			const command = new ListTasksCommand();
-
-			expect((command as any).validateOptions({ format: 'invalid' })).toBe(
-				false
-			);
-			expect(consoleErrorSpy).toHaveBeenCalledWith(
-				expect.stringContaining('Invalid format: invalid')
-			);
-
-			consoleErrorSpy.mockRestore();
-		});
-	});
-});

apps/cli/tests/unit/commands/show.command.spec.ts

@@ -5,10 +5,14 @@
 import type { TmCore } from '@tm/core';
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 
-// Mock dependencies
-vi.mock('@tm/core', () => ({
-	createTmCore: vi.fn()
-}));
+// Mock dependencies - use partial mock to keep TaskIdSchema and other exports
+vi.mock('@tm/core', async (importOriginal) => {
+	const actual = await importOriginal<typeof import('@tm/core')>();
+	return {
+		...actual,
+		createTmCore: vi.fn()
+	};
+});
 
 vi.mock('../../../src/utils/project-root.js', () => ({
 	getProjectRoot: vi.fn((path?: string) => path || '/test/project')

apps/docs/capabilities/rpg-method.mdx

@@ -317,7 +317,7 @@ Without test strategy, the AI won't know what to test during implementation.
 
 - [PRD Creation and Parsing Guide](/getting-started/quick-start/prd-quick)
 - [Task Structure Documentation](/capabilities/task-structure)
-- [Microsoft Research RPG Paper](https://arxiv.org/abs/2410.21376) (Original methodology)
+- [Microsoft Research RPG Paper](https://arxiv.org/html/2509.16198) (Original methodology)
 
 ---
 

apps/mcp/package.json

@@ -17,8 +17,8 @@
 		"test": "vitest run",
 		"test:watch": "vitest",
 		"test:coverage": "vitest run --coverage",
-		"test:unit": "vitest run -t unit",
-		"test:integration": "vitest run -t integration",
+		"test:unit": "vitest run '**/*.spec.ts'",
+		"test:integration": "vitest run '**/*.test.ts'",
 		"test:ci": "vitest run --coverage --reporter=dot"
 	},
 	"dependencies": {

apps/mcp/tests/integration/tools/generate.tool.test.ts

@@ -158,7 +158,7 @@ describe('generate MCP tool', () => {
 		const response = callMCPTool('generate', { projectRoot: testDir });
 
 		expect(response.data.orphanedFilesRemoved).toBe(1);
-	}, 15000);
+	}, 30000); // Longer timeout: this test makes 2 MCP calls (generate, then regenerate)
 
 	it('should accept output parameter for custom directory', () => {
 		const testData = createTasksFile({

docs/models.md

@@ -1,4 +1,4 @@
-# Available Models as of December 18, 2025
+# Available Models as of January 15, 2026
 
 ## Main Models
 
@@ -15,10 +15,9 @@
 | claude-code | opus                                           | 0.725     | 0          | 0           |
 | claude-code | sonnet                                         | 0.727     | 0          | 0           |
 | claude-code | haiku                                          | 0.45      | 0          | 0           |
-| codex-cli   | gpt-5                                          | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5-codex                                    | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5.1                                        | 0.76      | 0          | 0           |
+| codex-cli   | gpt-5.2-codex                                  | 0.82      | 0          | 0           |
 | codex-cli   | gpt-5.1-codex-max                              | 0.78      | 0          | 0           |
+| codex-cli   | gpt-5.1-codex-mini                             | 0.72      | 0          | 0           |
 | codex-cli   | gpt-5.2                                        | 0.8       | 0          | 0           |
 | mcp         | mcp-sampling                                   | —         | 0          | 0           |
 | gemini-cli  | gemini-3-flash-preview                         | —         | 0          | 0           |
@@ -132,10 +131,9 @@
 | claude-code | opus                                         | 0.725     | 0          | 0           |
 | claude-code | sonnet                                       | 0.727     | 0          | 0           |
 | claude-code | haiku                                        | 0.45      | 0          | 0           |
-| codex-cli   | gpt-5                                        | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5-codex                                  | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5.1                                      | 0.76      | 0          | 0           |
+| codex-cli   | gpt-5.2-codex                                | 0.82      | 0          | 0           |
 | codex-cli   | gpt-5.1-codex-max                            | 0.78      | 0          | 0           |
+| codex-cli   | gpt-5.1-codex-mini                           | 0.72      | 0          | 0           |
 | codex-cli   | gpt-5.2                                      | 0.8       | 0          | 0           |
 | mcp         | mcp-sampling                                 | —         | 0          | 0           |
 | gemini-cli  | gemini-3-flash-preview                       | —         | 0          | 0           |
@@ -191,10 +189,9 @@
 | claude-code | opus                                           | 0.725     | 0          | 0           |
 | claude-code | sonnet                                         | 0.727     | 0          | 0           |
 | claude-code | haiku                                          | 0.45      | 0          | 0           |
-| codex-cli   | gpt-5                                          | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5-codex                                    | 0.749     | 0          | 0           |
-| codex-cli   | gpt-5.1                                        | 0.76      | 0          | 0           |
+| codex-cli   | gpt-5.2-codex                                  | 0.82      | 0          | 0           |
 | codex-cli   | gpt-5.1-codex-max                              | 0.78      | 0          | 0           |
+| codex-cli   | gpt-5.1-codex-mini                             | 0.72      | 0          | 0           |
 | codex-cli   | gpt-5.2                                        | 0.8       | 0          | 0           |
 | mcp         | mcp-sampling                                   | —         | 0          | 0           |
 | gemini-cli  | gemini-3-flash-preview                         | —         | 0          | 0           |

package-lock.json

@@ -1,12 +1,12 @@
 {
 	"name": "task-master-ai",
-	"version": "0.41.0",
+	"version": "0.40.1",
 	"lockfileVersion": 3,
 	"requires": true,
 	"packages": {
 		"": {
 			"name": "task-master-ai",
-			"version": "0.41.0",
+			"version": "0.40.1",
 			"license": "MIT WITH Commons-Clause",
 			"workspaces": [
 				"apps/*",
@@ -67,6 +67,7 @@
 				"ollama-ai-provider-v2": "^1.3.1",
 				"open": "^10.2.0",
 				"ora": "^8.2.0",
+				"proper-lockfile": "^4.1.2",
 				"simple-git": "^3.28.0",
 				"steno": "^4.0.2",
 				"terminal-link": "^5.0.0",
@@ -13713,6 +13714,16 @@
 			"integrity": "sha512-PIzZZlEppgrpoT2QgbnDU+MMzuR6BbCjllj0bM70lWoejMeNJAxCchxnv7J3XFkI8MpygtRpzXrIlmWUBclP5A==",
 			"license": "MIT"
 		},
+		"node_modules/@types/proper-lockfile": {
+			"version": "4.1.4",
+			"resolved": "https://registry.npmjs.org/@types/proper-lockfile/-/proper-lockfile-4.1.4.tgz",
+			"integrity": "sha512-uo2ABllncSqg9F1D4nugVl9v93RmjxF6LJzQLMLDdPaXCUIDPeOJ21Gbqi43xNKzBi/WQ0Q0dICqufzQbMjipQ==",
+			"dev": true,
+			"license": "MIT",
+			"dependencies": {
+				"@types/retry": "*"
+			}
+		},
 		"node_modules/@types/react": {
 			"version": "19.1.8",
 			"resolved": "https://registry.npmjs.org/@types/react/-/react-19.1.8.tgz",
@@ -13783,6 +13794,13 @@
 				"node": ">= 0.6"
 			}
 		},
+		"node_modules/@types/retry": {
+			"version": "0.12.5",
+			"resolved": "https://registry.npmjs.org/@types/retry/-/retry-0.12.5.tgz",
+			"integrity": "sha512-3xSjTp3v03X/lSQLkczaN9UIEwJMoMCA1+Nb5HfbJEQWogdeQIyVtTvxPXDQjZ5zws8rFQfVfRdz03ARihPJgw==",
+			"dev": true,
+			"license": "MIT"
+		},
 		"node_modules/@types/stack-utils": {
 			"version": "2.0.3",
 			"resolved": "https://registry.npmjs.org/@types/stack-utils/-/stack-utils-2.0.3.tgz",
@@ -29634,6 +29652,23 @@
 				"node": ">= 6"
 			}
 		},
+		"node_modules/proper-lockfile": {
+			"version": "4.1.2",
+			"resolved": "https://registry.npmjs.org/proper-lockfile/-/proper-lockfile-4.1.2.tgz",
+			"integrity": "sha512-TjNPblN4BwAWMXU8s9AEz4JmQxnD1NNL7bNOY/AKUzyamc379FWASUhc/K1pL2noVb+XmZKLL68cjzLsiOAMaA==",
+			"license": "MIT",
+			"dependencies": {
+				"graceful-fs": "^4.2.4",
+				"retry": "^0.12.0",
+				"signal-exit": "^3.0.2"
+			}
+		},
+		"node_modules/proper-lockfile/node_modules/signal-exit": {
+			"version": "3.0.7",
+			"resolved": "https://registry.npmjs.org/signal-exit/-/signal-exit-3.0.7.tgz",
+			"integrity": "sha512-wnD2ZE+l+SPC/uoS0vXeE9L1+0wuaMqKlfz9AMUo38JsyLSBWSFcHR1Rri62LZc12vLr1gb3jl7iwQhgwpAbGQ==",
+			"license": "ISC"
+		},
 		"node_modules/property-information": {
 			"version": "7.1.0",
 			"resolved": "https://registry.npmjs.org/property-information/-/property-information-7.1.0.tgz",
@@ -31111,6 +31146,15 @@
 				"url": "https://opencollective.com/unified"
 			}
 		},
+		"node_modules/retry": {
+			"version": "0.12.0",
+			"resolved": "https://registry.npmjs.org/retry/-/retry-0.12.0.tgz",
+			"integrity": "sha512-9LkiTwjUh6rT555DtE9rTX+BKByPfrMzEAtnlEtdEwr3Nkffwiihqe2bWADg+OQRjt9gl6ICdmB/ZFDCGAtSow==",
+			"license": "MIT",
+			"engines": {
+				"node": ">= 4"
+			}
+		},
 		"node_modules/retry-request": {
 			"version": "7.0.2",
 			"resolved": "https://registry.npmjs.org/retry-request/-/retry-request-7.0.2.tgz",
@@ -36324,6 +36368,7 @@
 				"@supabase/supabase-js": "^2.57.4",
 				"date-fns": "^4.1.0",
 				"fs-extra": "^11.3.2",
+				"proper-lockfile": "^4.1.2",
 				"simple-git": "^3.28.0",
 				"steno": "^4.0.2",
 				"zod": "^4.1.11"
@@ -36331,6 +36376,7 @@
 			"devDependencies": {
 				"@types/fs-extra": "^11.0.4",
 				"@types/node": "^22.10.5",
+				"@types/proper-lockfile": "^4.1.4",
 				"@vitest/coverage-v8": "^4.0.10",
 				"strip-literal": "3.1.0",
 				"typescript": "^5.9.2",

package.json

@@ -20,6 +20,9 @@
 		"turbo:dev": "turbo dev",
 		"turbo:build": "turbo build",
 		"turbo:typecheck": "turbo typecheck",
+		"turbo:test": "turbo test",
+		"turbo:test:unit": "turbo test:unit",
+		"turbo:test:integration": "turbo test:integration",
 		"build:build-config": "npm run build -w @tm/build-config",
 		"test": "cross-env NODE_ENV=test node --experimental-vm-modules node_modules/.bin/jest",
 		"test:unit": "node --experimental-vm-modules node_modules/.bin/jest --testPathPattern=unit",
@@ -117,7 +120,8 @@
 		"turndown": "^7.2.2",
 		"undici": "^7.16.0",
 		"uuid": "^11.1.0",
-		"zod": "^4.1.12"
+		"zod": "^4.1.12",
+		"proper-lockfile": "^4.1.2"
 	},
 	"optionalDependencies": {
 		"@anthropic-ai/claude-code": "^2.0.59",

packages/tm-core/package.json

@@ -21,6 +21,8 @@
 	},
 	"scripts": {
 		"test": "vitest run",
+		"test:unit": "vitest run '**/*.spec.ts'",
+		"test:integration": "vitest run '**/*.test.ts'",
 		"test:watch": "vitest",
 		"test:coverage": "vitest run --coverage",
 		"lint": "biome check --write",
@@ -34,12 +36,14 @@
 		"@supabase/supabase-js": "^2.57.4",
 		"date-fns": "^4.1.0",
 		"fs-extra": "^11.3.2",
+		"proper-lockfile": "^4.1.2",
 		"simple-git": "^3.28.0",
 		"steno": "^4.0.2",
 		"zod": "^4.1.11"
 	},
 	"devDependencies": {
 		"@types/fs-extra": "^11.0.4",
+		"@types/proper-lockfile": "^4.1.4",
 		"@types/node": "^22.10.5",
 		"@vitest/coverage-v8": "^4.0.10",
 		"strip-literal": "3.1.0",

packages/tm-core/src/index.ts

@@ -204,6 +204,17 @@ export {
 	type GenerateTaskFilesResult
 } from './modules/tasks/services/task-file-generator.service.js';
 
+// Task filtering utilities
+export {
+	buildBlocksMap,
+	filterReadyTasks,
+	filterBlockingTasks,
+	ACTIONABLE_STATUSES,
+	type TaskWithBlocks,
+	type InvalidDependency,
+	type BuildBlocksMapResult
+} from './modules/tasks/utils/index.js';
+
 // Integration - Advanced
 export {
 	ExportService,

packages/tm-core/src/modules/config/managers/config-manager.spec.ts

@@ -34,72 +34,64 @@ describe('ConfigManager', () => {
 			}
 		});
 
-		// Setup default mock behaviors
-		vi.mocked(ConfigLoader).mockImplementation(
-			() =>
-				({
-					getDefaultConfig: vi.fn().mockReturnValue({
-						models: { main: 'default-model', fallback: 'fallback-model' },
-						storage: { type: 'file' },
-						version: '1.0.0'
-					}),
-					loadLocalConfig: vi.fn().mockResolvedValue(null),
-					loadGlobalConfig: vi.fn().mockResolvedValue(null),
-					hasLocalConfig: vi.fn().mockResolvedValue(false),
-					hasGlobalConfig: vi.fn().mockResolvedValue(false)
-				}) as any
-		);
+		// Setup default mock behaviors using class syntax for proper constructor mocking
+		vi.mocked(ConfigLoader).mockImplementation(function (this: any) {
+			this.getDefaultConfig = vi.fn().mockReturnValue({
+				models: { main: 'default-model', fallback: 'fallback-model' },
+				storage: { type: 'file' },
+				version: '1.0.0'
+			});
+			this.loadLocalConfig = vi.fn().mockResolvedValue(null);
+			this.loadGlobalConfig = vi.fn().mockResolvedValue(null);
+			this.hasLocalConfig = vi.fn().mockResolvedValue(false);
+			this.hasGlobalConfig = vi.fn().mockResolvedValue(false);
+			return this;
+		} as any);
 
-		vi.mocked(ConfigMerger).mockImplementation(
-			() =>
-				({
-					addSource: vi.fn(),
-					clearSources: vi.fn(),
-					merge: vi.fn().mockReturnValue({
-						models: { main: 'merged-model', fallback: 'fallback-model' },
-						storage: { type: 'file' }
-					}),
-					getSources: vi.fn().mockReturnValue([]),
-					hasSource: vi.fn().mockReturnValue(false),
-					removeSource: vi.fn().mockReturnValue(false)
-				}) as any
-		);
+		vi.mocked(ConfigMerger).mockImplementation(function (this: any) {
+			this.addSource = vi.fn();
+			this.clearSources = vi.fn();
+			this.merge = vi.fn().mockReturnValue({
+				models: { main: 'merged-model', fallback: 'fallback-model' },
+				storage: { type: 'file' }
+			});
+			this.getSources = vi.fn().mockReturnValue([]);
+			this.hasSource = vi.fn().mockReturnValue(false);
+			this.removeSource = vi.fn().mockReturnValue(false);
+			return this;
+		} as any);
 
-		vi.mocked(RuntimeStateManager).mockImplementation(
-			() =>
-				({
-					loadState: vi.fn().mockResolvedValue({ activeTag: 'master' }),
-					saveState: vi.fn().mockResolvedValue(undefined),
-					getCurrentTag: vi.fn().mockReturnValue('master'),
-					setCurrentTag: vi.fn().mockResolvedValue(undefined),
-					getState: vi.fn().mockReturnValue({ activeTag: 'master' }),
-					updateMetadata: vi.fn().mockResolvedValue(undefined),
-					clearState: vi.fn().mockResolvedValue(undefined)
-				}) as any
-		);
+		vi.mocked(RuntimeStateManager).mockImplementation(function (this: any) {
+			this.loadState = vi.fn().mockResolvedValue({ activeTag: 'master' });
+			this.saveState = vi.fn().mockResolvedValue(undefined);
+			this.getCurrentTag = vi.fn().mockReturnValue('master');
+			this.setCurrentTag = vi.fn().mockResolvedValue(undefined);
+			this.getState = vi.fn().mockReturnValue({ activeTag: 'master' });
+			this.updateMetadata = vi.fn().mockResolvedValue(undefined);
+			this.clearState = vi.fn().mockResolvedValue(undefined);
+			return this;
+		} as any);
 
-		vi.mocked(ConfigPersistence).mockImplementation(
-			() =>
-				({
-					saveConfig: vi.fn().mockResolvedValue(undefined),
-					configExists: vi.fn().mockResolvedValue(false),
-					deleteConfig: vi.fn().mockResolvedValue(undefined),
-					getBackups: vi.fn().mockResolvedValue([]),
-					restoreFromBackup: vi.fn().mockResolvedValue(undefined)
-				}) as any
-		);
+		vi.mocked(ConfigPersistence).mockImplementation(function (this: any) {
+			this.saveConfig = vi.fn().mockResolvedValue(undefined);
+			this.configExists = vi.fn().mockResolvedValue(false);
+			this.deleteConfig = vi.fn().mockResolvedValue(undefined);
+			this.getBackups = vi.fn().mockResolvedValue([]);
+			this.restoreFromBackup = vi.fn().mockResolvedValue(undefined);
+			return this;
+		} as any);
 
-		vi.mocked(EnvironmentConfigProvider).mockImplementation(
-			() =>
-				({
-					loadConfig: vi.fn().mockReturnValue({}),
-					getRuntimeState: vi.fn().mockReturnValue({}),
-					hasEnvVar: vi.fn().mockReturnValue(false),
-					getAllTaskmasterEnvVars: vi.fn().mockReturnValue({}),
-					addMapping: vi.fn(),
-					getMappings: vi.fn().mockReturnValue([])
-				}) as any
-		);
+		vi.mocked(EnvironmentConfigProvider).mockImplementation(function (
+			this: any
+		) {
+			this.loadConfig = vi.fn().mockReturnValue({});
+			this.getRuntimeState = vi.fn().mockReturnValue({});
+			this.hasEnvVar = vi.fn().mockReturnValue(false);
+			this.getAllTaskmasterEnvVars = vi.fn().mockReturnValue({});
+			this.addMapping = vi.fn();
+			this.getMappings = vi.fn().mockReturnValue([]);
+			return this;
+		} as any);
 
 		// Since constructor is private, we need to use the factory method
 		// But for testing, we'll create a test instance using create()
@@ -178,28 +170,30 @@ describe('ConfigManager', () => {
 
 		it('should return storage configuration', () => {
 			const storage = manager.getStorageConfig();
-			expect(storage).toEqual({ type: 'file' });
+			expect(storage).toEqual({
+				type: 'file',
+				basePath: testProjectRoot,
+				apiConfigured: false
+			});
 		});
 
 		it('should return API storage configuration when configured', async () => {
 			// Create a new instance with API storage config
-			vi.mocked(ConfigMerger).mockImplementationOnce(
-				() =>
-					({
-						addSource: vi.fn(),
-						clearSources: vi.fn(),
-						merge: vi.fn().mockReturnValue({
-							storage: {
-								type: 'api',
-								apiEndpoint: 'https://api.example.com',
-								apiAccessToken: 'token123'
-							}
-						}),
-						getSources: vi.fn().mockReturnValue([]),
-						hasSource: vi.fn().mockReturnValue(false),
-						removeSource: vi.fn().mockReturnValue(false)
-					}) as any
-			);
+			vi.mocked(ConfigMerger).mockImplementationOnce(function (this: any) {
+				this.addSource = vi.fn();
+				this.clearSources = vi.fn();
+				this.merge = vi.fn().mockReturnValue({
+					storage: {
+						type: 'api',
+						apiEndpoint: 'https://api.example.com',
+						apiAccessToken: 'token123'
+					}
+				});
+				this.getSources = vi.fn().mockReturnValue([]);
+				this.hasSource = vi.fn().mockReturnValue(false);
+				this.removeSource = vi.fn().mockReturnValue(false);
+				return this;
+			} as any);
 
 			const apiManager = await ConfigManager.create(testProjectRoot);
 
@@ -207,7 +201,9 @@ describe('ConfigManager', () => {
 			expect(storage).toEqual({
 				type: 'api',
 				apiEndpoint: 'https://api.example.com',
-				apiAccessToken: 'token123'
+				apiAccessToken: 'token123',
+				basePath: testProjectRoot,
+				apiConfigured: true
 			});
 		});
 

packages/tm-core/src/modules/config/services/config-loader.service.spec.ts

@@ -2,17 +2,12 @@
  * @fileoverview Unit tests for ConfigLoader service
  */
 
-import fs from 'node:fs/promises';
+import * as fsPromises from 'node:fs/promises';
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 import { DEFAULT_CONFIG_VALUES } from '../../../common/interfaces/configuration.interface.js';
 import { ConfigLoader } from './config-loader.service.js';
 
-vi.mock('node:fs', () => ({
-	promises: {
-		readFile: vi.fn(),
-		access: vi.fn()
-	}
-}));
+vi.mock('node:fs/promises');
 
 describe('ConfigLoader', () => {
 	let configLoader: ConfigLoader;
@@ -56,11 +51,13 @@ describe('ConfigLoader', () => {
 				storage: { type: 'api' as const }
 			};
 
-			vi.mocked(fs.readFile).mockResolvedValue(JSON.stringify(mockConfig));
+			vi.mocked(fsPromises.readFile).mockResolvedValue(
+				JSON.stringify(mockConfig)
+			);
 
 			const result = await configLoader.loadLocalConfig();
 
-			expect(fs.readFile).toHaveBeenCalledWith(
+			expect(fsPromises.readFile).toHaveBeenCalledWith(
 				'/test/project/.taskmaster/config.json',
 				'utf-8'
 			);
@@ -70,7 +67,7 @@ describe('ConfigLoader', () => {
 		it('should return null when config file does not exist', async () => {
 			const error = new Error('File not found') as any;
 			error.code = 'ENOENT';
-			vi.mocked(fs.readFile).mockRejectedValue(error);
+			vi.mocked(fsPromises.readFile).mockRejectedValue(error);
 
 			const result = await configLoader.loadLocalConfig();
 
@@ -79,7 +76,7 @@ describe('ConfigLoader', () => {
 
 		it('should throw TaskMasterError for other file errors', async () => {
 			const error = new Error('Permission denied');
-			vi.mocked(fs.readFile).mockRejectedValue(error);
+			vi.mocked(fsPromises.readFile).mockRejectedValue(error);
 
 			await expect(configLoader.loadLocalConfig()).rejects.toThrow(
 				'Failed to load local configuration'
@@ -87,7 +84,7 @@ describe('ConfigLoader', () => {
 		});
 
 		it('should throw error for invalid JSON', async () => {
-			vi.mocked(fs.readFile).mockResolvedValue('invalid json');
+			vi.mocked(fsPromises.readFile).mockResolvedValue('invalid json');
 
 			await expect(configLoader.loadLocalConfig()).rejects.toThrow();
 		});
@@ -102,18 +99,18 @@ describe('ConfigLoader', () => {
 
 	describe('hasLocalConfig', () => {
 		it('should return true when local config exists', async () => {
-			vi.mocked(fs.access).mockResolvedValue(undefined);
+			vi.mocked(fsPromises.access).mockResolvedValue(undefined);
 
 			const result = await configLoader.hasLocalConfig();
 
-			expect(fs.access).toHaveBeenCalledWith(
+			expect(fsPromises.access).toHaveBeenCalledWith(
 				'/test/project/.taskmaster/config.json'
 			);
 			expect(result).toBe(true);
 		});
 
 		it('should return false when local config does not exist', async () => {
-			vi.mocked(fs.access).mockRejectedValue(new Error('Not found'));
+			vi.mocked(fsPromises.access).mockRejectedValue(new Error('Not found'));
 
 			const result = await configLoader.hasLocalConfig();
 
@@ -123,18 +120,18 @@ describe('ConfigLoader', () => {
 
 	describe('hasGlobalConfig', () => {
 		it('should check global config path', async () => {
-			vi.mocked(fs.access).mockResolvedValue(undefined);
+			vi.mocked(fsPromises.access).mockResolvedValue(undefined);
 
 			const result = await configLoader.hasGlobalConfig();
 
-			expect(fs.access).toHaveBeenCalledWith(
+			expect(fsPromises.access).toHaveBeenCalledWith(
 				expect.stringContaining('.taskmaster/config.json')
 			);
 			expect(result).toBe(true);
 		});
 
 		it('should return false when global config does not exist', async () => {
-			vi.mocked(fs.access).mockRejectedValue(new Error('Not found'));
+			vi.mocked(fsPromises.access).mockRejectedValue(new Error('Not found'));
 
 			const result = await configLoader.hasGlobalConfig();
 

packages/tm-core/src/modules/config/services/runtime-state-manager.service.spec.ts

@@ -2,19 +2,12 @@
  * @fileoverview Unit tests for RuntimeStateManager service
  */
 
-import fs from 'node:fs/promises';
+import * as fs from 'node:fs/promises';
 import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest';
 import { DEFAULT_CONFIG_VALUES } from '../../../common/interfaces/configuration.interface.js';
 import { RuntimeStateManager } from './runtime-state-manager.service.js';
 
-vi.mock('node:fs', () => ({
-	promises: {
-		readFile: vi.fn(),
-		writeFile: vi.fn(),
-		mkdir: vi.fn(),
-		unlink: vi.fn()
-	}
-}));
+vi.mock('node:fs/promises');
 
 describe('RuntimeStateManager', () => {
 	let stateManager: RuntimeStateManager;
@@ -96,15 +89,10 @@ describe('RuntimeStateManager', () => {
 		it('should handle invalid JSON gracefully', async () => {
 			vi.mocked(fs.readFile).mockResolvedValue('invalid json');
 
-			// Mock console.warn to avoid noise in tests
-			const warnSpy = vi.spyOn(console, 'warn').mockImplementation(() => {});
-
 			const state = await stateManager.loadState();
 
+			// Should return default state when JSON is invalid
 			expect(state.currentTag).toBe(DEFAULT_CONFIG_VALUES.TAGS.DEFAULT_TAG);
-			expect(warnSpy).toHaveBeenCalled();
-
-			warnSpy.mockRestore();
 		});
 	});
 
@@ -124,7 +112,7 @@ describe('RuntimeStateManager', () => {
 			// Verify writeFile was called with correct data
 			expect(fs.writeFile).toHaveBeenCalledWith(
 				'/test/project/.taskmaster/state.json',
-				expect.stringContaining('"activeTag":"test-tag"'),
+				expect.stringContaining('"currentTag": "test-tag"'),
 				'utf-8'
 			);
 

packages/tm-core/src/modules/loop/loop-domain.spec.ts

@@ -102,7 +102,7 @@ describe('LoopDomain', () => {
 				tag: 'feature-branch'
 			};
 			const config = (loopDomain as any).buildConfig(fullConfig);
-			expect(config).toEqual(fullConfig);
+			expect(config).toMatchObject(fullConfig);
 		});
 	});
 

packages/tm-core/src/modules/loop/loop-domain.ts

@@ -31,9 +31,9 @@ export class LoopDomain {
 
 	/**
 	 * Check if Docker sandbox auth is ready
-	 * @returns true if ready, false if auth needed
+	 * @returns Object with ready status and optional error message
 	 */
-	checkSandboxAuth(): boolean {
+	checkSandboxAuth(): { ready: boolean; error?: string } {
 		const service = new LoopService({ projectRoot: this.projectRoot });
 		return service.checkSandboxAuth();
 	}
@@ -41,10 +41,11 @@ export class LoopDomain {
 	/**
 	 * Run Docker sandbox session for user authentication
 	 * Blocks until user completes auth
+	 * @returns Object with success status and optional error message
 	 */
-	runInteractiveAuth(): void {
+	runInteractiveAuth(): { success: boolean; error?: string } {
 		const service = new LoopService({ projectRoot: this.projectRoot });
-		service.runInteractiveAuth();
+		return service.runInteractiveAuth();
 	}
 
 	// ========== Loop Operations ==========
@@ -188,7 +189,8 @@ export class LoopDomain {
 				partial.progressFile ??
 				path.join(this.projectRoot, '.taskmaster', 'progress.txt'),
 			sleepSeconds: partial.sleepSeconds ?? 5,
-			tag: partial.tag
+			tag: partial.tag,
+			sandbox: partial.sandbox ?? false
 		};
 	}
 }

packages/tm-core/src/modules/loop/services/loop.service.spec.ts

@@ -122,7 +122,7 @@ describe('LoopService', () => {
 	});
 
 	describe('checkSandboxAuth()', () => {
-		it('should return true when output contains ok', () => {
+		it('should return ready=true when output contains ok', () => {
 			mockSpawnSync.mockReturnValue({
 				stdout: 'OK',
 				stderr: '',
@@ -135,7 +135,7 @@ describe('LoopService', () => {
 			const service = new LoopService(defaultOptions);
 			const result = service.checkSandboxAuth();
 
-			expect(result).toBe(true);
+			expect(result.ready).toBe(true);
 			expect(mockSpawnSync).toHaveBeenCalledWith(
 				'docker',
 				['sandbox', 'run', 'claude', '-p', 'Say OK'],
@@ -146,7 +146,7 @@ describe('LoopService', () => {
 			);
 		});
 
-		it('should return false when output does not contain ok', () => {
+		it('should return ready=false when output does not contain ok', () => {
 			mockSpawnSync.mockReturnValue({
 				stdout: 'Error: not authenticated',
 				stderr: '',
@@ -159,7 +159,7 @@ describe('LoopService', () => {
 			const service = new LoopService(defaultOptions);
 			const result = service.checkSandboxAuth();
 
-			expect(result).toBe(false);
+			expect(result.ready).toBe(false);
 		});
 
 		it('should check stderr as well as stdout', () => {
@@ -175,7 +175,7 @@ describe('LoopService', () => {
 			const service = new LoopService(defaultOptions);
 			const result = service.checkSandboxAuth();
 
-			expect(result).toBe(true);
+			expect(result.ready).toBe(true);
 		});
 	});
 
@@ -256,7 +256,7 @@ describe('LoopService', () => {
 				expect(mockSpawnSync).toHaveBeenCalledTimes(3);
 			});
 
-			it('should call spawnSync with docker sandbox run claude -p', async () => {
+			it('should call spawnSync with claude -p by default (non-sandbox)', async () => {
 				mockSpawnSync.mockReturnValue({
 					stdout: 'Done',
 					stderr: '',
@@ -274,14 +274,8 @@ describe('LoopService', () => {
 				});
 
 				expect(mockSpawnSync).toHaveBeenCalledWith(
-					'docker',
-					expect.arrayContaining([
-						'sandbox',
-						'run',
-						'claude',
-						'-p',
-						expect.any(String)
-					]),
+					'claude',
+					expect.arrayContaining(['-p', expect.any(String)]),
 					expect.objectContaining({
 						cwd: '/test/project'
 					})
@@ -397,7 +391,8 @@ describe('LoopService', () => {
 				expect(fsPromises.mkdir).toHaveBeenCalledWith('/test', {
 					recursive: true
 				});
-				expect(fsPromises.writeFile).toHaveBeenCalledWith(
+				// Uses appendFile instead of writeFile to preserve existing progress
+				expect(fsPromises.appendFile).toHaveBeenCalledWith(
 					'/test/progress.txt',
 					expect.stringContaining('# Task Master Loop Progress'),
 					'utf-8'
@@ -449,8 +444,8 @@ describe('LoopService', () => {
 
 				// Verify spawn was called with prompt containing iteration info
 				const spawnCall = mockSpawnSync.mock.calls[0];
-				// Args are ['sandbox', 'run', 'claude', '-p', prompt]
-				const promptArg = spawnCall[1][4];
+				// Args are ['-p', prompt, '--dangerously-skip-permissions'] for non-sandbox
+				const promptArg = spawnCall[1][1];
 				expect(promptArg).toContain('iteration 1 of 1');
 			});
 

packages/tm-core/src/modules/loop/services/loop.service.ts

@@ -1,9 +1,9 @@
 /**
- * @fileoverview Loop Service - Orchestrates running Claude Code in Docker sandbox iterations
+ * @fileoverview Loop Service - Orchestrates running Claude Code iterations (sandbox or CLI mode)
  */
 
 import { spawnSync } from 'node:child_process';
-import { appendFile, mkdir, readFile, writeFile } from 'node:fs/promises';
+import { appendFile, mkdir, readFile } from 'node:fs/promises';
 import path from 'node:path';
 import { PRESETS, isPreset as checkIsPreset } from '../presets/index.js';
 import type {
@@ -34,7 +34,7 @@ export class LoopService {
 	}
 
 	/** Check if Docker sandbox auth is ready */
-	checkSandboxAuth(): boolean {
+	checkSandboxAuth(): { ready: boolean; error?: string } {
 		const result = spawnSync(
 			'docker',
 			['sandbox', 'run', 'claude', '-p', 'Say OK'],
@@ -42,16 +42,29 @@ export class LoopService {
 				cwd: this.projectRoot,
 				timeout: 30000,
 				encoding: 'utf-8',
-				stdio: ['inherit', 'pipe', 'pipe'] // stdin from terminal, capture stdout/stderr
+				stdio: ['inherit', 'pipe', 'pipe']
 			}
 		);
+
+		if (result.error) {
+			const code = (result.error as NodeJS.ErrnoException).code;
+			if (code === 'ENOENT') {
+				return {
+					ready: false,
+					error:
+						'Docker is not installed. Install Docker Desktop to use --sandbox mode.'
+				};
+			}
+			return { ready: false, error: `Docker error: ${result.error.message}` };
+		}
+
 		const output = (result.stdout || '') + (result.stderr || '');
-		return output.toLowerCase().includes('ok');
+		return { ready: output.toLowerCase().includes('ok') };
 	}
 
 	/** Run interactive Docker sandbox session for user authentication */
-	runInteractiveAuth(): void {
-		spawnSync(
+	runInteractiveAuth(): { success: boolean; error?: string } {
+		const result = spawnSync(
 			'docker',
 			[
 				'sandbox',
@@ -64,6 +77,34 @@ export class LoopService {
 				stdio: 'inherit'
 			}
 		);
+
+		if (result.error) {
+			const code = (result.error as NodeJS.ErrnoException).code;
+			if (code === 'ENOENT') {
+				return {
+					success: false,
+					error:
+						'Docker is not installed. Install Docker Desktop to use --sandbox mode.'
+				};
+			}
+			return { success: false, error: `Docker error: ${result.error.message}` };
+		}
+
+		if (result.status === null) {
+			return {
+				success: false,
+				error: 'Docker terminated abnormally (no exit code)'
+			};
+		}
+
+		if (result.status !== 0) {
+			return {
+				success: false,
+				error: `Docker exited with code ${result.status}`
+			};
+		}
+
+		return { success: true };
 	}
 
 	/** Run a loop with the given configuration */
@@ -80,7 +121,11 @@ export class LoopService {
 			console.log(`━━━ Iteration ${i} of ${config.iterations} ━━━`);
 
 			const prompt = await this.buildPrompt(config, i);
-			const iteration = this.executeIteration(prompt, i);
+			const iteration = this.executeIteration(
+				prompt,
+				i,
+				config.sandbox ?? false
+			);
 			iterations.push(iteration);
 
 			// Check for early exit conditions
@@ -135,9 +180,11 @@ export class LoopService {
 	private async initProgressFile(config: LoopConfig): Promise<void> {
 		await mkdir(path.dirname(config.progressFile), { recursive: true });
 		const tagLine = config.tag ? `# Tag: ${config.tag}\n` : '';
-		await writeFile(
+		// Append to existing progress file instead of overwriting
+		await appendFile(
 			config.progressFile,
-			`# Task Master Loop Progress
+			`
+# Task Master Loop Progress
 # Started: ${new Date().toISOString()}
 # Preset: ${config.prompt}
 # Max Iterations: ${config.iterations}
@@ -217,30 +264,64 @@ Loop iteration ${iteration} of ${config.iterations}${tagInfo}`;
 
 	private executeIteration(
 		prompt: string,
-		iterationNum: number
+		iterationNum: number,
+		sandbox: boolean
 	): LoopIteration {
 		const startTime = Date.now();
 
-		const result = spawnSync(
-			'docker',
-			['sandbox', 'run', 'claude', '-p', prompt],
-			{
-				cwd: this.projectRoot,
-				encoding: 'utf-8',
-				maxBuffer: 50 * 1024 * 1024, // 50MB buffer
-				stdio: ['inherit', 'pipe', 'pipe']
+		// Use docker sandbox or plain claude based on config
+		const command = sandbox ? 'docker' : 'claude';
+		const args = sandbox
+			? ['sandbox', 'run', 'claude', '-p', prompt]
+			: ['-p', prompt, '--dangerously-skip-permissions'];
+
+		const result = spawnSync(command, args, {
+			cwd: this.projectRoot,
+			encoding: 'utf-8',
+			maxBuffer: 50 * 1024 * 1024, // 50MB buffer
+			stdio: ['inherit', 'pipe', 'pipe']
+		});
+
+		// Check for spawn-level errors (command not found, permission denied, etc.)
+		if (result.error) {
+			const code = (result.error as NodeJS.ErrnoException).code;
+			let errorMessage: string;
+
+			if (code === 'ENOENT') {
+				errorMessage = sandbox
+					? 'Docker is not installed. Install Docker Desktop to use --sandbox mode.'
+					: 'Claude CLI is not installed. Install with: npm install -g @anthropic-ai/claude-code';
+			} else if (code === 'EACCES') {
+				errorMessage = `Permission denied executing '${command}'`;
+			} else {
+				errorMessage = `Failed to execute '${command}': ${result.error.message}`;
 			}
-		);
+
+			console.error(`[Loop Error] ${errorMessage}`);
+			return {
+				iteration: iterationNum,
+				status: 'error',
+				duration: Date.now() - startTime,
+				message: errorMessage
+			};
+		}
 
 		const output = (result.stdout || '') + (result.stderr || '');
 
 		// Print output to console (spawnSync with pipe captures but doesn't display)
 		if (output) console.log(output);
 
-		const { status, message } = this.parseCompletion(
-			output,
-			result.status ?? 1
-		);
+		// Handle null status (spawn failed but no error object - shouldn't happen but be safe)
+		if (result.status === null) {
+			return {
+				iteration: iterationNum,
+				status: 'error',
+				duration: Date.now() - startTime,
+				message: 'Command terminated abnormally (no exit code)'
+			};
+		}
+
+		const { status, message } = this.parseCompletion(output, result.status);
 		return {
 			iteration: iterationNum,
 			status,

packages/tm-core/src/modules/loop/types.ts

@@ -26,6 +26,8 @@ export interface LoopConfig {
 	sleepSeconds: number;
 	/** Tag context to operate on (optional) */
 	tag?: string;
+	/** Run Claude in Docker sandbox mode (default: false) */
+	sandbox?: boolean;
 }
 
 /**

packages/tm-core/src/modules/storage/adapters/file-storage/file-operations.spec.ts

@@ -0,0 +1,264 @@
+/**
+ * Tests for FileOperations class
+ * Focuses on modifyJson and cross-process locking functionality
+ */
+
+import { describe, it, expect, beforeEach, afterEach } from 'vitest';
+import fs from 'node:fs/promises';
+import fsSync from 'node:fs';
+import path from 'node:path';
+import os from 'node:os';
+import { FileOperations } from './file-operations.js';
+
+describe('FileOperations', () => {
+	let tempDir: string;
+	let testFilePath: string;
+	let fileOps: FileOperations;
+
+	beforeEach(async () => {
+		// Create a temp directory for each test
+		tempDir = fsSync.mkdtempSync(path.join(os.tmpdir(), 'tm-core-test-'));
+		testFilePath = path.join(tempDir, 'test.json');
+		fileOps = new FileOperations();
+	});
+
+	afterEach(async () => {
+		// Clean up
+		await fileOps.cleanup();
+		if (tempDir && fsSync.existsSync(tempDir)) {
+			fsSync.rmSync(tempDir, { recursive: true, force: true });
+		}
+	});
+
+	describe('modifyJson', () => {
+		it('should modify existing JSON data', async () => {
+			// Set up initial data
+			await fs.writeFile(testFilePath, JSON.stringify({ count: 0 }));
+
+			// Modify data
+			await fileOps.modifyJson(testFilePath, (data: { count: number }) => ({
+				...data,
+				count: data.count + 1
+			}));
+
+			// Verify
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.count).toBe(1);
+		});
+
+		it('should create file if it does not exist', async () => {
+			const newFilePath = path.join(tempDir, 'new-file.json');
+
+			await fileOps.modifyJson(newFilePath, () => ({ created: true }));
+
+			expect(fsSync.existsSync(newFilePath)).toBe(true);
+			const result = JSON.parse(await fs.readFile(newFilePath, 'utf-8'));
+			expect(result.created).toBe(true);
+		});
+
+		it('should handle async modifier functions', async () => {
+			await fs.writeFile(testFilePath, JSON.stringify({ value: 'initial' }));
+
+			await fileOps.modifyJson(
+				testFilePath,
+				async (data: { value: string }) => {
+					// Simulate async operation
+					await new Promise((resolve) => setTimeout(resolve, 10));
+					return { ...data, value: 'modified' };
+				}
+			);
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.value).toBe('modified');
+		});
+
+		it('should re-read file inside lock to prevent stale data', async () => {
+			// Initial data
+			await fs.writeFile(testFilePath, JSON.stringify({ version: 1 }));
+
+			// Simulate two sequential modifications
+			await fileOps.modifyJson(testFilePath, (data: { version: number }) => ({
+				version: data.version + 1
+			}));
+
+			await fileOps.modifyJson(testFilePath, (data: { version: number }) => ({
+				version: data.version + 1
+			}));
+
+			// Both modifications should have been applied
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.version).toBe(3);
+		});
+
+		it('should not leave lock files on success', async () => {
+			await fs.writeFile(testFilePath, JSON.stringify({}));
+
+			await fileOps.modifyJson(testFilePath, (data) => ({
+				...data,
+				modified: true
+			}));
+
+			// Check no lock files exist
+			const files = await fs.readdir(tempDir);
+			const lockFiles = files.filter((f) => f.endsWith('.lock'));
+			expect(lockFiles).toHaveLength(0);
+		});
+
+		it('should release lock even if modifier throws', async () => {
+			await fs.writeFile(testFilePath, JSON.stringify({}));
+
+			await expect(
+				fileOps.modifyJson(testFilePath, () => {
+					throw new Error('Modifier error');
+				})
+			).rejects.toThrow('Modifier error');
+
+			// Should still be able to acquire lock for another operation
+			await fileOps.modifyJson(testFilePath, () => ({ recovered: true }));
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.recovered).toBe(true);
+		});
+
+		it('should handle empty file gracefully', async () => {
+			// Create empty file
+			await fs.writeFile(testFilePath, '');
+
+			await fileOps.modifyJson(testFilePath, () => ({ initialized: true }));
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.initialized).toBe(true);
+		});
+
+		it('should handle file with only whitespace', async () => {
+			await fs.writeFile(testFilePath, '   \n  ');
+
+			await fileOps.modifyJson(testFilePath, () => ({ initialized: true }));
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.initialized).toBe(true);
+		});
+
+		it('should throw on corrupted JSON', async () => {
+			// Write invalid JSON that is not empty
+			await fs.writeFile(testFilePath, '{ invalid json content');
+
+			await expect(
+				fileOps.modifyJson(testFilePath, (data) => data)
+			).rejects.toThrow(/Corrupted JSON/);
+		});
+
+		it('should preserve complex nested structures', async () => {
+			const complexData = {
+				tasks: [
+					{
+						id: 1,
+						title: 'Task 1',
+						subtasks: [{ id: '1.1', title: 'Subtask' }]
+					}
+				],
+				metadata: {
+					created: '2024-01-01',
+					tags: ['tag1', 'tag2']
+				}
+			};
+			await fs.writeFile(testFilePath, JSON.stringify(complexData, null, 2));
+
+			await fileOps.modifyJson(testFilePath, (data: typeof complexData) => ({
+				...data,
+				tasks: [...data.tasks, { id: 2, title: 'Task 2', subtasks: [] }]
+			}));
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.tasks).toHaveLength(2);
+			expect(result.tasks[0].subtasks).toHaveLength(1);
+			expect(result.metadata.tags).toEqual(['tag1', 'tag2']);
+		});
+	});
+
+	describe('concurrent operations', () => {
+		it('should serialize truly concurrent modifyJson calls', async () => {
+			// Initial data
+			await fs.writeFile(testFilePath, JSON.stringify({ count: 0 }));
+
+			const numConcurrentWrites = 5;
+			const writes = [];
+
+			for (let i = 0; i < numConcurrentWrites; i++) {
+				writes.push(
+					fileOps.modifyJson(testFilePath, (data: { count: number }) => ({
+						count: data.count + 1
+					}))
+				);
+			}
+
+			await Promise.all(writes);
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.count).toBe(numConcurrentWrites);
+		});
+
+		it('should handle concurrent writes from multiple FileOperations instances', async () => {
+			// Initial data
+			await fs.writeFile(testFilePath, JSON.stringify({ count: 0 }));
+
+			const numInstances = 3;
+			const instances = Array.from(
+				{ length: numInstances },
+				() => new FileOperations()
+			);
+			const writes = instances.map((ops) =>
+				ops.modifyJson(testFilePath, (data: { count: number }) => ({
+					count: data.count + 1
+				}))
+			);
+
+			await Promise.all(writes);
+
+			// Cleanup all instances
+			await Promise.all(instances.map((ops) => ops.cleanup()));
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.count).toBe(numInstances);
+		});
+	});
+
+	describe('writeJson', () => {
+		it('should write JSON atomically', async () => {
+			const data = { test: 'value' };
+
+			await fileOps.writeJson(testFilePath, data);
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.test).toBe('value');
+		});
+
+		it('should not leave temp files on success', async () => {
+			await fileOps.writeJson(testFilePath, { test: true });
+
+			const files = await fs.readdir(tempDir);
+			const tempFiles = files.filter((f) => f.includes('.tmp'));
+			expect(tempFiles).toHaveLength(0);
+
+			// Also verify no lock files remain
+			const lockFiles = files.filter((f) => f.endsWith('.lock'));
+			expect(lockFiles).toHaveLength(0);
+		});
+	});
+
+	describe('cleanup', () => {
+		it('should clear cached writers', async () => {
+			// Write to create a cached writer
+			await fileOps.writeJson(testFilePath, { test: 1 });
+
+			// Cleanup
+			await fileOps.cleanup();
+
+			// Should still work after cleanup (creates new writer)
+			await fileOps.writeJson(testFilePath, { test: 2 });
+
+			const result = JSON.parse(await fs.readFile(testFilePath, 'utf-8'));
+			expect(result.test).toBe(2);
+		});
+	});
+});

packages/tm-core/src/modules/storage/adapters/file-storage/file-operations.ts

@@ -1,16 +1,52 @@
 /**
- * @fileoverview File operations with atomic writes and locking
+ * @fileoverview File operations with atomic writes and cross-process locking
+ *
+ * Uses steno for atomic writes (same pattern as workflow-state-manager.ts)
+ * and proper-lockfile for cross-process locking to prevent lost updates.
  */
 
 import { constants } from 'node:fs';
 import fs from 'node:fs/promises';
+import path from 'node:path';
+import lockfile from 'proper-lockfile';
+import { Writer } from 'steno';
 import type { FileStorageData } from './format-handler.js';
 
 /**
- * Handles atomic file operations with locking mechanism
+ * File locking configuration for cross-process safety
+ */
+const LOCK_OPTIONS = {
+	stale: 10000, // Consider lock stale after 10 seconds
+	retries: {
+		retries: 5,
+		factor: 2,
+		minTimeout: 100,
+		maxTimeout: 1000
+	},
+	realpath: false // Don't resolve symlinks (faster)
+};
+
+/**
+ * Handles atomic file operations with cross-process locking mechanism.
+ *
+ * Writers are cached for reuse. Call {@link cleanup} when disposing of
+ * long-lived instances to prevent memory leaks.
  */
 export class FileOperations {
-	private fileLocks: Map<string, Promise<void>> = new Map();
+	/** Map of file paths to steno Writers for reuse */
+	private writers = new Map<string, Writer>();
+
+	/**
+	 * Get or create a steno Writer for a file path
+	 */
+	private getWriter(filePath: string): Writer {
+		let writer = this.writers.get(filePath);
+		if (!writer) {
+			writer = new Writer(filePath);
+			this.writers.set(filePath, writer);
+		}
+		return writer;
+	}
 
 	/**
 	 * Read and parse JSON file
@@ -31,52 +67,132 @@ export class FileOperations {
 	}
 
 	/**
-	 * Write JSON file with atomic operation and locking
+	 * Write JSON file with atomic operation and cross-process locking.
+	 * Uses steno for atomic writes and proper-lockfile for cross-process safety.
+	 * WARNING: This replaces the entire file. For concurrent modifications,
+	 * use modifyJson() instead to prevent lost updates.
 	 */
 	async writeJson(
 		filePath: string,
 		data: FileStorageData | any
 	): Promise<void> {
-		// Use file locking to prevent concurrent writes
-		const lockKey = filePath;
-		const existingLock = this.fileLocks.get(lockKey);
-
-		if (existingLock) {
-			await existingLock;
-		}
-
-		const lockPromise = this.performAtomicWrite(filePath, data);
-		this.fileLocks.set(lockKey, lockPromise);
+		// Ensure file exists for locking (proper-lockfile requires this)
+		await this.ensureFileExists(filePath);
 
+		// Acquire cross-process lock
+		let release: (() => Promise<void>) | null = null;
 		try {
-			await lockPromise;
+			release = await lockfile.lock(filePath, LOCK_OPTIONS);
+
+			// Use steno Writer for atomic writes (same pattern as workflow-state-manager)
+			const content = JSON.stringify(data, null, 2);
+			const writer = this.getWriter(filePath);
+			await writer.write(content);
 		} finally {
-			this.fileLocks.delete(lockKey);
+			if (release) {
+				try {
+					await release();
+				} catch (err: any) {
+					// Log but don't throw - lock may have been released already
+					// Other errors should be visible for debugging
+					if (process.env.DEBUG || process.env.TASKMASTER_DEBUG === 'true') {
+						console.warn(
+							`[WARN] Lock release warning for ${filePath}: ${err.message}`
+						);
+					}
+				}
+			}
 		}
 	}
 
 	/**
-	 * Perform atomic write operation using temporary file
+	 * Read-modify-write JSON file with cross-process locking.
+	 * Uses steno for atomic writes and proper-lockfile for cross-process safety.
+	 * Re-reads file inside lock to prevent lost updates from stale snapshots.
+	 * @param filePath - Path to the JSON file
+	 * @param modifier - Function that receives current data and returns modified data
 	 */
-	private async performAtomicWrite(filePath: string, data: any): Promise<void> {
-		const tempPath = `${filePath}.tmp`;
+	async modifyJson<T = any>(
+		filePath: string,
+		modifier: (currentData: T) => T | Promise<T>
+	): Promise<void> {
+		// Ensure file exists for locking (proper-lockfile requires this)
+		await this.ensureFileExists(filePath);
 
+		// Acquire cross-process lock
+		let release: (() => Promise<void>) | null = null;
 		try {
-			// Write to temp file first
-			const content = JSON.stringify(data, null, 2);
-			await fs.writeFile(tempPath, content, 'utf-8');
+			release = await lockfile.lock(filePath, LOCK_OPTIONS);
 
-			// Atomic rename
-			await fs.rename(tempPath, filePath);
-		} catch (error: any) {
-			// Clean up temp file if it exists
+			// Re-read file INSIDE lock to get current state
+			// This prevents lost updates from stale snapshots
+			let currentData: T;
 			try {
-				await fs.unlink(tempPath);
-			} catch {
-				// Ignore cleanup errors
+				const content = await fs.readFile(filePath, 'utf-8');
+				currentData = JSON.parse(content);
+			} catch (err: any) {
+				// Distinguish between expected empty/new files and actual corruption
+				if (err.code === 'ENOENT') {
+					// File doesn't exist yet - start fresh
+					currentData = {} as T;
+				} else if (err instanceof SyntaxError) {
+					// Check if it's just an empty file (our ensureFileExists writes '{}')
+					const content = await fs.readFile(filePath, 'utf-8').catch(() => '');
+					if (content.trim() === '' || content.trim() === '{}') {
+						currentData = {} as T;
+					} else {
+						// Actual JSON corruption - this is a serious error
+						throw new Error(
+							`Corrupted JSON in ${filePath}: ${err.message}. File contains: ${content.substring(0, 100)}...`
+						);
+					}
+				} else {
+					// Other errors (permission, I/O) should be surfaced
+					throw new Error(
+						`Failed to read ${filePath} for modification: ${err.message}`
+					);
+				}
 			}
 
-			throw new Error(`Failed to write file ${filePath}: ${error.message}`);
+			// Apply modification
+			const newData = await modifier(currentData);
+
+			// Write atomically using steno (same pattern as workflow-state-manager)
+			const content = JSON.stringify(newData, null, 2);
+			const writer = this.getWriter(filePath);
+			await writer.write(content);
+		} finally {
+			if (release) {
+				try {
+					await release();
+				} catch (err: any) {
+					// Log but don't throw - lock may have been released already
+					// Other errors should be visible for debugging
+					if (process.env.DEBUG || process.env.TASKMASTER_DEBUG === 'true') {
+						console.warn(
+							`[WARN] Lock release warning for ${filePath}: ${err.message}`
+						);
+					}
+				}
+			}
+		}
+	}
+
+	/**
+	 * Ensure file exists for locking (proper-lockfile requires the file to exist).
+	 * Uses atomic creation with 'wx' flag to prevent TOCTOU race conditions.
+	 */
+	private async ensureFileExists(filePath: string): Promise<void> {
+		const dir = path.dirname(filePath);
+		await fs.mkdir(dir, { recursive: true });
+		try {
+			// Use 'wx' flag for atomic create - fails if file exists (prevents race)
+			await fs.writeFile(filePath, '{}', { flag: 'wx' });
+		} catch (err: any) {
+			// EEXIST is expected if another process created the file - that's fine
+			if (err.code !== 'EEXIST') {
+				throw err;
+			}
 		}
 	}
 
@@ -159,13 +275,14 @@ export class FileOperations {
 	}
 
 	/**
-	 * Clean up all pending file operations
+	 * Clean up resources - releases cached steno Writers
+	 * Call this when the FileOperations instance is no longer needed
+	 * to prevent memory leaks in long-running processes.
 	 */
 	async cleanup(): Promise<void> {
-		const locks = Array.from(this.fileLocks.values());
-		if (locks.length > 0) {
-			await Promise.all(locks);
-		}
-		this.fileLocks.clear();
+		// Clear cached Writers to allow garbage collection
+		// Note: steno Writers don't have explicit close methods;
+		// they handle file descriptor cleanup internally
+		this.writers.clear();
 	}
 }

packages/tm-core/src/modules/storage/adapters/file-storage/file-storage.ts

@@ -220,6 +220,7 @@ export class FileStorage implements IStorage {
 
 	/**
 	 * Save tasks for a specific tag in the single tasks.json file
+	 * Uses modifyJson for atomic read-modify-write to prevent lost updates
 	 */
 	async saveTasks(tasks: Task[], tag?: string): Promise<void> {
 		const filePath = this.pathResolver.getTasksPath();
@@ -228,17 +229,6 @@ export class FileStorage implements IStorage {
 		// Ensure directory exists
 		await this.fileOps.ensureDir(this.pathResolver.getTasksDir());
 
-		// Get existing data from the file
-		let existingData: any = {};
-		try {
-			existingData = await this.fileOps.readJson(filePath);
-		} catch (error: any) {
-			if (error.code !== 'ENOENT') {
-				throw new Error(`Failed to read existing tasks: ${error.message}`);
-			}
-			// File doesn't exist, start with empty data
-		}
-
 		// Create metadata for this tag
 		const metadata: TaskMetadata = {
 			version: '1.0.0',
@@ -251,43 +241,44 @@ export class FileStorage implements IStorage {
 		// Normalize tasks
 		const normalizedTasks = this.normalizeTaskIds(tasks);
 
-		// Update the specific tag in the existing data structure
-		if (
-			this.formatHandler.detectFormat(existingData) === 'legacy' ||
-			Object.keys(existingData).some(
-				(key) => key !== 'tasks' && key !== 'metadata'
-			)
-		) {
-			// Legacy format - update/add the tag
-			existingData[resolvedTag] = {
-				tasks: normalizedTasks,
-				metadata
-			};
-		} else if (resolvedTag === 'master') {
-			// Standard format for master tag
-			existingData = {
-				tasks: normalizedTasks,
-				metadata
-			};
-		} else {
-			// Convert to legacy format when adding non-master tags
-			const masterTasks = existingData.tasks || [];
-			const masterMetadata = existingData.metadata || metadata;
-
-			existingData = {
-				master: {
-					tasks: masterTasks,
-					metadata: masterMetadata
-				},
-				[resolvedTag]: {
+		// Use modifyJson for atomic read-modify-write
+		await this.fileOps.modifyJson(filePath, (existingData: any) => {
+			// Update the specific tag in the existing data structure
+			if (
+				this.formatHandler.detectFormat(existingData) === 'legacy' ||
+				Object.keys(existingData).some(
+					(key) => key !== 'tasks' && key !== 'metadata'
+				)
+			) {
+				// Legacy format - update/add the tag
+				existingData[resolvedTag] = {
 					tasks: normalizedTasks,
 					metadata
-				}
-			};
-		}
+				};
+				return existingData;
+			} else if (resolvedTag === 'master') {
+				// Standard format for master tag
+				return {
+					tasks: normalizedTasks,
+					metadata
+				};
+			} else {
+				// Convert to legacy format when adding non-master tags
+				const masterTasks = existingData.tasks || [];
+				const masterMetadata = existingData.metadata || metadata;
 
-		// Write the updated file
-		await this.fileOps.writeJson(filePath, existingData);
+				return {
+					master: {
+						tasks: masterTasks,
+						metadata: masterMetadata
+					},
+					[resolvedTag]: {
+						tasks: normalizedTasks,
+						metadata
+					}
+				};
+			}
+		});
 	}
 
 	/**
@@ -594,6 +585,7 @@ export class FileStorage implements IStorage {
 
 	/**
 	 * Create a new tag in the tasks.json file
+	 * Uses modifyJson for atomic read-modify-write to prevent lost updates
 	 */
 	async createTag(
 		tagName: string,
@@ -602,62 +594,33 @@ export class FileStorage implements IStorage {
 		const filePath = this.pathResolver.getTasksPath();
 
 		try {
-			const existingData = await this.fileOps.readJson(filePath);
-			const format = this.formatHandler.detectFormat(existingData);
+			await this.fileOps.modifyJson(filePath, (existingData: any) => {
+				const format = this.formatHandler.detectFormat(existingData);
 
-			if (format === 'legacy') {
-				// Legacy format - add new tag key
-				if (tagName in existingData) {
-					throw new TaskMasterError(
-						`Tag ${tagName} already exists`,
-						ERROR_CODES.VALIDATION_ERROR
-					);
-				}
-
-				// Get tasks to copy if specified
-				let tasksToCopy = [];
-				if (options?.copyFrom) {
-					if (
-						options.copyFrom in existingData &&
-						existingData[options.copyFrom].tasks
-					) {
-						tasksToCopy = JSON.parse(
-							JSON.stringify(existingData[options.copyFrom].tasks)
+				if (format === 'legacy') {
+					// Legacy format - add new tag key
+					if (tagName in existingData) {
+						throw new TaskMasterError(
+							`Tag ${tagName} already exists`,
+							ERROR_CODES.VALIDATION_ERROR
 						);
 					}
-				}
 
-				// Create new tag structure
-				existingData[tagName] = {
-					tasks: tasksToCopy,
-					metadata: {
-						created: new Date().toISOString(),
-						updatedAt: new Date().toISOString(),
-						description:
-							options?.description ||
-							`Tag created on ${new Date().toLocaleDateString()}`,
-						tags: [tagName]
+					// Get tasks to copy if specified
+					let tasksToCopy: any[] = [];
+					if (options?.copyFrom) {
+						if (
+							options.copyFrom in existingData &&
+							existingData[options.copyFrom].tasks
+						) {
+							tasksToCopy = JSON.parse(
+								JSON.stringify(existingData[options.copyFrom].tasks)
+							);
+						}
 					}
-				};
 
-				await this.fileOps.writeJson(filePath, existingData);
-			} else {
-				// Standard format - need to convert to legacy format first
-				const masterTasks = existingData.tasks || [];
-				const masterMetadata = existingData.metadata || {};
-
-				// Get tasks to copy (from master in this case)
-				let tasksToCopy = [];
-				if (options?.copyFrom === 'master' || !options?.copyFrom) {
-					tasksToCopy = JSON.parse(JSON.stringify(masterTasks));
-				}
-
-				const newData = {
-					master: {
-						tasks: masterTasks,
-						metadata: { ...masterMetadata, tags: ['master'] }
-					},
-					[tagName]: {
+					// Create new tag structure
+					existingData[tagName] = {
 						tasks: tasksToCopy,
 						metadata: {
 							created: new Date().toISOString(),
@@ -667,11 +630,39 @@ export class FileStorage implements IStorage {
 								`Tag created on ${new Date().toLocaleDateString()}`,
 							tags: [tagName]
 						}
-					}
-				};
+					};
 
-				await this.fileOps.writeJson(filePath, newData);
-			}
+					return existingData;
+				} else {
+					// Standard format - need to convert to legacy format first
+					const masterTasks = existingData.tasks || [];
+					const masterMetadata = existingData.metadata || {};
+
+					// Get tasks to copy (from master in this case)
+					let tasksToCopy: any[] = [];
+					if (options?.copyFrom === 'master' || !options?.copyFrom) {
+						tasksToCopy = JSON.parse(JSON.stringify(masterTasks));
+					}
+
+					return {
+						master: {
+							tasks: masterTasks,
+							metadata: { ...masterMetadata, tags: ['master'] }
+						},
+						[tagName]: {
+							tasks: tasksToCopy,
+							metadata: {
+								created: new Date().toISOString(),
+								updatedAt: new Date().toISOString(),
+								description:
+									options?.description ||
+									`Tag created on ${new Date().toLocaleDateString()}`,
+								tags: [tagName]
+							}
+						}
+					};
+				}
+			});
 		} catch (error: any) {
 			if (error.code === 'ENOENT') {
 				throw new Error('Tasks file not found - initialize project first');
@@ -682,26 +673,41 @@ export class FileStorage implements IStorage {
 
 	/**
 	 * Delete a tag from the single tasks.json file
+	 * Uses modifyJson for atomic read-modify-write to prevent lost updates
 	 */
 	async deleteTag(tag: string): Promise<void> {
 		const filePath = this.pathResolver.getTasksPath();
 
 		try {
-			const existingData = await this.fileOps.readJson(filePath);
+			// Use modifyJson to handle all cases atomically
+			let shouldDeleteFile = false;
 
-			if (this.formatHandler.detectFormat(existingData) === 'legacy') {
-				// Legacy format - remove the tag key
-				if (tag in existingData) {
-					delete existingData[tag];
-					await this.fileOps.writeJson(filePath, existingData);
-				} else {
-					throw new Error(`Tag ${tag} not found`);
+			await this.fileOps.modifyJson(filePath, (data: any) => {
+				if (
+					this.formatHandler.detectFormat(data) !== 'legacy' &&
+					tag === 'master'
+				) {
+					// Standard format - mark for file deletion after lock release
+					shouldDeleteFile = true;
+					return data; // Return unchanged, we'll delete the file after
 				}
-			} else if (tag === 'master') {
-				// Standard format - delete the entire file for master tag
+
+				if (this.formatHandler.detectFormat(data) === 'legacy') {
+					// Legacy format - remove the tag key
+					if (tag in data) {
+						delete data[tag];
+						return data;
+					} else {
+						throw new Error(`Tag ${tag} not found`);
+					}
+				} else {
+					throw new Error(`Tag ${tag} not found in standard format`);
+				}
+			});
+
+			// Delete the file if we're removing master tag from standard format
+			if (shouldDeleteFile) {
 				await this.fileOps.deleteFile(filePath);
-			} else {
-				throw new Error(`Tag ${tag} not found in standard format`);
 			}
 		} catch (error: any) {
 			if (error.code === 'ENOENT') {
@@ -713,44 +719,43 @@ export class FileStorage implements IStorage {
 
 	/**
 	 * Rename a tag within the single tasks.json file
+	 * Uses modifyJson for atomic read-modify-write to prevent lost updates
 	 */
 	async renameTag(oldTag: string, newTag: string): Promise<void> {
 		const filePath = this.pathResolver.getTasksPath();
 
 		try {
-			const existingData = await this.fileOps.readJson(filePath);
+			await this.fileOps.modifyJson(filePath, (existingData: any) => {
+				if (this.formatHandler.detectFormat(existingData) === 'legacy') {
+					// Legacy format - rename the tag key
+					if (oldTag in existingData) {
+						existingData[newTag] = existingData[oldTag];
+						delete existingData[oldTag];
 
-			if (this.formatHandler.detectFormat(existingData) === 'legacy') {
-				// Legacy format - rename the tag key
-				if (oldTag in existingData) {
-					existingData[newTag] = existingData[oldTag];
-					delete existingData[oldTag];
+						// Update metadata tags array
+						if (existingData[newTag].metadata) {
+							existingData[newTag].metadata.tags = [newTag];
+						}
 
-					// Update metadata tags array
-					if (existingData[newTag].metadata) {
-						existingData[newTag].metadata.tags = [newTag];
+						return existingData;
+					} else {
+						throw new Error(`Tag ${oldTag} not found`);
 					}
+				} else if (oldTag === 'master') {
+					// Convert standard format to legacy when renaming master
+					const masterTasks = existingData.tasks || [];
+					const masterMetadata = existingData.metadata || {};
 
-					await this.fileOps.writeJson(filePath, existingData);
+					return {
+						[newTag]: {
+							tasks: masterTasks,
+							metadata: { ...masterMetadata, tags: [newTag] }
+						}
+					};
 				} else {
-					throw new Error(`Tag ${oldTag} not found`);
+					throw new Error(`Tag ${oldTag} not found in standard format`);
 				}
-			} else if (oldTag === 'master') {
-				// Convert standard format to legacy when renaming master
-				const masterTasks = existingData.tasks || [];
-				const masterMetadata = existingData.metadata || {};
-
-				const newData = {
-					[newTag]: {
-						tasks: masterTasks,
-						metadata: { ...masterMetadata, tags: [newTag] }
-					}
-				};
-
-				await this.fileOps.writeJson(filePath, newData);
-			} else {
-				throw new Error(`Tag ${oldTag} not found in standard format`);
-			}
+			});
 		} catch (error: any) {
 			if (error.code === 'ENOENT') {
 				throw new Error(`Tag ${oldTag} not found - file doesn't exist`);

packages/tm-core/src/modules/tasks/utils/index.ts

@@ -0,0 +1,6 @@
+/**
+ * @fileoverview Task utility exports
+ * Re-exports task filtering and analysis utilities
+ */
+
+export * from './task-filters.js';

packages/tm-core/src/modules/tasks/utils/task-filters.ts

@@ -0,0 +1,168 @@
+/**
+ * @fileoverview Task filtering utilities for dependency and readiness analysis
+ * Business logic for filtering tasks by actionable status, dependencies, and blocking relationships
+ */
+
+import type { Task, TaskStatus } from '../../../common/types/index.js';
+import {
+	TASK_STATUSES,
+	isTaskComplete
+} from '../../../common/constants/index.js';
+import { getLogger } from '../../../common/logger/index.js';
+
+const logger = getLogger('TaskFilters');
+
+/**
+ * Task with blocks field (inverse of dependencies)
+ * A task's blocks array contains IDs of tasks that depend on it
+ */
+export type TaskWithBlocks = Task & { blocks: string[] };
+
+/**
+ * Statuses that are actionable (not deferred, blocked, or terminal)
+ * Tasks with these statuses can be worked on when dependencies are satisfied
+ */
+export const ACTIONABLE_STATUSES: readonly TaskStatus[] = [
+	'pending',
+	'in-progress',
+	'review'
+] as const;
+
+/**
+ * Invalid dependency reference (task depends on non-existent task)
+ */
+export interface InvalidDependency {
+	/** ID of the task with the invalid dependency */
+	taskId: string;
+	/** ID of the non-existent dependency */
+	depId: string;
+}
+
+/**
+ * Result of building the blocks map with validation information
+ */
+export interface BuildBlocksMapResult {
+	/** Map of task ID -> array of task IDs that depend on it */
+	blocksMap: Map<string, string[]>;
+	/** Array of invalid dependency references (dependencies to non-existent tasks) */
+	invalidDependencies: InvalidDependency[];
+}
+
+/**
+ * Build a map of task ID -> array of task IDs that depend on it (blocks)
+ * This is the inverse of the dependencies relationship
+ *
+ * Also validates dependencies and returns any references to non-existent tasks.
+ *
+ * @param tasks - Array of tasks to analyze
+ * @returns Object containing the blocks map and any invalid dependency references
+ *
+ * @example
+ * ```typescript
+ * const tasks = [
+ *   { id: '1', dependencies: [] },
+ *   { id: '2', dependencies: ['1'] },
+ *   { id: '3', dependencies: ['1', '2'] }
+ * ];
+ * const { blocksMap, invalidDependencies } = buildBlocksMap(tasks);
+ * // blocksMap.get('1') => ['2', '3']  // Task 1 blocks tasks 2 and 3
+ * // blocksMap.get('2') => ['3']       // Task 2 blocks task 3
+ * // blocksMap.get('3') => []          // Task 3 blocks nothing
+ * // invalidDependencies => []         // No invalid deps in this example
+ * ```
+ */
+export function buildBlocksMap(tasks: Task[]): BuildBlocksMapResult {
+	const blocksMap = new Map<string, string[]>(
+		tasks.map((task) => [String(task.id), []])
+	);
+	const invalidDependencies: InvalidDependency[] = [];
+
+	// For each task, add it to the blocks list of each of its dependencies
+	for (const task of tasks) {
+		for (const depId of task.dependencies ?? []) {
+			const depIdStr = String(depId);
+			const blocks = blocksMap.get(depIdStr);
+			if (blocks) {
+				blocks.push(String(task.id));
+			} else {
+				// Dependency references a non-existent task
+				invalidDependencies.push({
+					taskId: String(task.id),
+					depId: depIdStr
+				});
+			}
+		}
+	}
+
+	return { blocksMap, invalidDependencies };
+}
+
+/**
+ * Filter to only tasks that are ready to work on
+ * A task is ready when:
+ * 1. It has an actionable status (pending, in-progress, or review)
+ * 2. All its dependencies are complete (done, completed, or cancelled)
+ *
+ * @param tasks - Array of tasks with blocks information
+ * @returns Filtered array of tasks that are ready to work on
+ *
+ * @example
+ * ```typescript
+ * const tasks = [
+ *   { id: '1', status: 'done', dependencies: [], blocks: ['2'] },
+ *   { id: '2', status: 'pending', dependencies: ['1'], blocks: [] },
+ *   { id: '3', status: 'pending', dependencies: ['2'], blocks: [] }
+ * ];
+ * const readyTasks = filterReadyTasks(tasks);
+ * // Returns only task 2: status is actionable and dependency '1' is done
+ * // Task 3 is not ready because dependency '2' is still pending
+ * ```
+ */
+export function filterReadyTasks(tasks: TaskWithBlocks[]): TaskWithBlocks[] {
+	// Build set of completed task IDs for dependency checking
+	const completedIds = new Set<string>(
+		tasks.filter((t) => isTaskComplete(t.status)).map((t) => String(t.id))
+	);
+
+	return tasks.filter((task) => {
+		// Validate status is a known value
+		if (!TASK_STATUSES.includes(task.status)) {
+			logger.warn(
+				`Task ${task.id} has unexpected status "${task.status}". Valid statuses are: ${TASK_STATUSES.join(', ')}`
+			);
+		}
+
+		// Must be in an actionable status (excludes deferred, blocked, done, cancelled)
+		if (!ACTIONABLE_STATUSES.includes(task.status)) {
+			return false;
+		}
+
+		// Ready if no dependencies or all dependencies are completed
+		const deps = task.dependencies ?? [];
+		return deps.every((depId) => completedIds.has(String(depId)));
+	});
+}
+
+/**
+ * Filter to only tasks that block other tasks
+ * These are tasks that have at least one other task depending on them
+ *
+ * @param tasks - Array of tasks with blocks information
+ * @returns Filtered array of tasks that have dependents (block other tasks)
+ *
+ * @example
+ * ```typescript
+ * const tasks = [
+ *   { id: '1', blocks: ['2', '3'] },  // Blocks tasks 2 and 3
+ *   { id: '2', blocks: [] },          // Blocks nothing
+ *   { id: '3', blocks: [] }           // Blocks nothing
+ * ];
+ * const blockingTasks = filterBlockingTasks(tasks);
+ * // Returns only task 1 (the only task with non-empty blocks)
+ * ```
+ */
+export function filterBlockingTasks(
+	tasks: TaskWithBlocks[]
+): TaskWithBlocks[] {
+	return tasks.filter((task) => task.blocks.length > 0);
+}

packages/tm-core/src/modules/tasks/validation/task-id.ts

@@ -6,14 +6,15 @@
  *
  * FILE STORAGE (local):
  * - Main tasks: "1", "2", "15"
- * - Subtasks: "1.2", "15.3" (one level only)
+ * - Subtasks: "1.2", "15.3"
+ * - Sub-subtasks: "1.2.3", "15.3.1"
  *
  * API STORAGE (Hamster):
  * - Main tasks: "HAM-1", "ham-1", "HAM1", "ham1" (all normalized to "HAM-1")
+ * - Variable-length prefixes: "PROJ-456", "TAS-1", "abc-999"
  * - No subtasks (API doesn't use dot notation)
  *
  * NOT supported:
- * - Deep nesting: "1.2.3" (file storage only has one subtask level)
  * - API subtasks: "HAM-1.2" (doesn't exist)
  */
 
@@ -24,10 +25,10 @@ import { normalizeDisplayId } from '../../../common/schemas/task-id.schema.js';
  * Pattern for validating a single task ID
  * Permissive input - accepts with or without hyphen for API IDs
  * - Numeric: "1", "15", "999"
- * - Numeric subtasks: "1.2" (one level only)
- * - API display IDs: "HAM-1", "ham-1", "HAM1", "ham1"
+ * - Numeric subtasks: "1.2", "1.2.3" (multi-level supported)
+ * - API display IDs: "HAM-1", "PROJ-456", "TAS-1", "abc-999"
  */
-export const TASK_ID_PATTERN = /^(\d+(\.\d+)?|[A-Za-z]{3}-?\d+)$/;
+export const TASK_ID_PATTERN = /^(\d+(\.\d+)*|[A-Za-z]+-?\d+)$/;
 
 /**
  * Validates a single task ID string
@@ -39,9 +40,10 @@ export const TASK_ID_PATTERN = /^(\d+(\.\d+)?|[A-Za-z]{3}-?\d+)$/;
  * ```typescript
  * isValidTaskIdFormat("1");        // true
  * isValidTaskIdFormat("1.2");      // true
+ * isValidTaskIdFormat("1.2.3");    // true (multi-level subtasks)
  * isValidTaskIdFormat("HAM-1");    // true
+ * isValidTaskIdFormat("PROJ-456"); // true (variable-length prefix)
  * isValidTaskIdFormat("ham1");     // true (permissive input)
- * isValidTaskIdFormat("1.2.3");    // false (too deep)
  * isValidTaskIdFormat("HAM-1.2");  // false (no API subtasks)
  * isValidTaskIdFormat("abc");      // false
  * ```

packages/tm-profiles/package.json

@@ -10,6 +10,8 @@
 	},
 	"scripts": {
 		"test": "vitest run",
+		"test:unit": "vitest run '**/*.spec.ts'",
+		"test:integration": "vitest run '**/*.test.ts'",
 		"test:watch": "vitest",
 		"test:coverage": "vitest run --coverage",
 		"lint": "biome check --write",

scripts/modules/supported-models.json

@@ -130,41 +130,21 @@
 	],
 	"codex-cli": [
 		{
-			"id": "gpt-5",
-			"swe_score": 0.749,
+			"id": "gpt-5.2-codex",
+			"name": "GPT-5.2 Codex",
+			"swe_score": 0.82,
 			"cost_per_1m_tokens": {
 				"input": 0,
 				"output": 0
 			},
 			"allowed_roles": ["main", "fallback", "research"],
 			"max_tokens": 128000,
-			"supported": true
-		},
-		{
-			"id": "gpt-5-codex",
-			"swe_score": 0.749,
-			"cost_per_1m_tokens": {
-				"input": 0,
-				"output": 0
-			},
-			"allowed_roles": ["main", "fallback", "research"],
-			"max_tokens": 128000,
-			"supported": true
-		},
-		{
-			"id": "gpt-5.1",
-			"swe_score": 0.76,
-			"cost_per_1m_tokens": {
-				"input": 0,
-				"output": 0
-			},
-			"allowed_roles": ["main", "fallback", "research"],
-			"max_tokens": 128000,
-			"reasoning_efforts": ["none", "low", "medium", "high"],
+			"reasoning_efforts": ["none", "low", "medium", "high", "xhigh"],
 			"supported": true
 		},
 		{
 			"id": "gpt-5.1-codex-max",
+			"name": "GPT-5.1 Codex Max",
 			"swe_score": 0.78,
 			"cost_per_1m_tokens": {
 				"input": 0,
@@ -175,8 +155,22 @@
 			"reasoning_efforts": ["none", "low", "medium", "high", "xhigh"],
 			"supported": true
 		},
+		{
+			"id": "gpt-5.1-codex-mini",
+			"name": "GPT-5.1 Codex Mini",
+			"swe_score": 0.72,
+			"cost_per_1m_tokens": {
+				"input": 0,
+				"output": 0
+			},
+			"allowed_roles": ["main", "fallback", "research"],
+			"max_tokens": 128000,
+			"reasoning_efforts": ["none", "low", "medium", "high"],
+			"supported": true
+		},
 		{
 			"id": "gpt-5.2",
+			"name": "GPT-5.2",
 			"swe_score": 0.8,
 			"cost_per_1m_tokens": {
 				"input": 0,

scripts/modules/task-manager/tag-management.js

@@ -10,6 +10,7 @@ import {
 	tryListTagsViaRemote,
 	tryUseTagViaRemote
 } from '@tm/bridge';
+import { filterReadyTasks, isTaskComplete } from '@tm/core';
 import { displayBanner, getStatusWithColor } from '../ui.js';
 import {
 	findProjectRoot,
@@ -531,6 +532,7 @@ async function enhanceTagsWithMetadata(tasksPath, rawData, context = {}) {
  * @param {Object} options - Options object
  * @param {boolean} [options.showTaskCounts=true] - Whether to show task counts
  * @param {boolean} [options.showMetadata=false] - Whether to show metadata
+ * @param {boolean} [options.ready=false] - Whether to filter to only tags with ready tasks
  * @param {Object} context - Context object containing session and projectRoot
  * @param {string} [context.projectRoot] - Project root path
  * @param {Object} [context.mcpLog] - MCP logger object (optional)
@@ -544,7 +546,11 @@ async function tags(
 	outputFormat = 'text'
 ) {
 	const { mcpLog, projectRoot } = context;
-	const { showTaskCounts = true, showMetadata = false } = options;
+	const {
+		showTaskCounts = true,
+		showMetadata = false,
+		ready = false
+	} = options;
 
 	// Create a consistent logFn object regardless of context
 	const logFn = mcpLog || {
@@ -619,12 +625,16 @@ async function tags(
 			const tasks = tagData.tasks || [];
 			const metadata = tagData.metadata || {};
 
+			// Use centralized filtering from @tm/core
+			// Note: filterReadyTasks expects TaskWithBlocks[] but only uses status/dependencies at runtime
+			const tasksWithBlocks = tasks.map((t) => ({ ...t, blocks: [] }));
+			const readyTasks = filterReadyTasks(tasksWithBlocks);
+
 			tagList.push({
 				name: tagName,
 				isCurrent: tagName === currentTag,
-				completedTasks: tasks.filter(
-					(t) => t.status === 'done' || t.status === 'completed'
-				).length,
+				completedTasks: tasks.filter((t) => isTaskComplete(t.status)).length,
+				readyTasks: readyTasks.length,
 				tasks: tasks || [],
 				created: metadata.created || 'Unknown',
 				description: metadata.description || 'No description'
@@ -638,22 +648,32 @@ async function tags(
 			return a.name.localeCompare(b.name);
 		});
 
-		logFn.success(`Found ${tagList.length} tags`);
+		// Filter to only tags with ready tasks if --ready flag is set
+		let filteredTagList = tagList;
+		if (ready) {
+			filteredTagList = tagList.filter((tag) => tag.readyTasks > 0);
+			logFn.info(`Filtered to ${filteredTagList.length} tags with ready tasks`);
+		}
+
+		logFn.success(`Found ${filteredTagList.length} tags`);
 
 		// For JSON output, return structured data
 		if (outputFormat === 'json') {
 			return {
-				tags: tagList,
+				tags: filteredTagList,
 				currentTag,
-				totalTags: tagList.length
+				totalTags: filteredTagList.length
 			};
 		}
 
 		// For text output, display formatted table
 		if (outputFormat === 'text') {
-			if (tagList.length === 0) {
+			if (filteredTagList.length === 0) {
+				const message = ready
+					? 'No tags with ready tasks found'
+					: 'No tags found';
 				console.log(
-					boxen(chalk.yellow('No tags found'), {
+					boxen(chalk.yellow(message), {
 						padding: 1,
 						borderColor: 'yellow',
 						borderStyle: 'round',
@@ -667,7 +687,8 @@ async function tags(
 			const headers = [chalk.cyan.bold('Tag Name')];
 			if (showTaskCounts) {
 				headers.push(chalk.cyan.bold('Tasks'));
-				headers.push(chalk.cyan.bold('Completed'));
+				headers.push(chalk.cyan.bold('Ready'));
+				headers.push(chalk.cyan.bold('Done'));
 			}
 			if (showMetadata) {
 				headers.push(chalk.cyan.bold('Created'));
@@ -680,16 +701,16 @@ async function tags(
 
 			let colWidths;
 			if (showMetadata) {
-				// With metadata: Tag Name, Tasks, Completed, Created, Description
-				const widths = [0.25, 0.1, 0.12, 0.15, 0.38];
+				// With metadata: Tag Name, Tasks, Ready, Done, Created, Description
+				const widths = [0.22, 0.08, 0.08, 0.08, 0.14, 0.38];
 				colWidths = widths.map((w, i) =>
-					Math.max(Math.floor(usableWidth * w), i === 0 ? 15 : 8)
+					Math.max(Math.floor(usableWidth * w), i === 0 ? 15 : 6)
 				);
 			} else {
-				// Without metadata: Tag Name, Tasks, Completed
-				const widths = [0.7, 0.15, 0.15];
+				// Without metadata: Tag Name, Tasks, Ready, Done
+				const widths = [0.6, 0.13, 0.13, 0.13];
 				colWidths = widths.map((w, i) =>
-					Math.max(Math.floor(usableWidth * w), i === 0 ? 20 : 10)
+					Math.max(Math.floor(usableWidth * w), i === 0 ? 20 : 8)
 				);
 			}
 
@@ -700,7 +721,7 @@ async function tags(
 			});
 
 			// Add rows
-			tagList.forEach((tag) => {
+			filteredTagList.forEach((tag) => {
 				const row = [];
 
 				// Tag name with current indicator
@@ -711,6 +732,11 @@ async function tags(
 
 				if (showTaskCounts) {
 					row.push(chalk.white(tag.tasks.length.toString()));
+					row.push(
+						tag.readyTasks > 0
+							? chalk.yellow(tag.readyTasks.toString())
+							: chalk.gray('0')
+					);
 					row.push(chalk.green(tag.completedTasks.toString()));
 				}
 
@@ -743,9 +769,9 @@ async function tags(
 		}
 
 		return {
-			tags: tagList,
+			tags: filteredTagList,
 			currentTag,
-			totalTags: tagList.length
+			totalTags: filteredTagList.length
 		};
 	} catch (error) {
 		logFn.error(`Error listing tags: ${error.message}`);

scripts/modules/utils.js

@@ -19,6 +19,278 @@ import * as gitUtils from './utils/git-utils.js';
 // Global silent mode flag
 let silentMode = false;
 
+// File locking configuration for cross-process safety
+const LOCK_CONFIG = {
+	maxRetries: 5,
+	retryDelay: 100, // ms
+	staleLockAge: 10000 // 10 seconds
+};
+
+/**
+ * Async sleep helper
+ */
+const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
+
+/**
+ * Synchronous sleep helper with Atomics.wait fallback
+ * Uses Atomics.wait when SharedArrayBuffer is available (proper non-busy wait),
+ * otherwise falls back to a busy-wait loop (less efficient but always works).
+ * @param {number} ms - Milliseconds to sleep
+ */
+function sleepSync(ms) {
+	// Check if SharedArrayBuffer and Atomics.wait are available
+	// They may not be available in some environments (e.g., browsers without proper headers)
+	if (
+		typeof SharedArrayBuffer !== 'undefined' &&
+		typeof Atomics !== 'undefined' &&
+		typeof Atomics.wait === 'function'
+	) {
+		try {
+			const sharedBuffer = new SharedArrayBuffer(4);
+			const int32 = new Int32Array(sharedBuffer);
+			Atomics.wait(int32, 0, 0, ms);
+			return;
+		} catch {
+			// Fall through to busy-wait fallback
+		}
+	}
+
+	// Fallback: busy-wait loop (less efficient but universally compatible)
+	// Note: This may cause high CPU usage for longer waits. Consider if this
+	// becomes an issue with large exponential backoff delays.
+	const end = Date.now() + ms;
+	while (Date.now() < end) {
+		// Busy wait - intentionally empty
+	}
+}
+
+/**
+ * Acquires an exclusive lock on a file and executes a callback
+ * Uses same lock file format as withFileLockSync for cross-process compatibility
+ * @param {string} filepath - Path to the file to lock
+ * @param {Function} callback - Async function to execute while holding the lock
+ * @param {Object} [options] - Options for lock behavior
+ * @param {boolean} [options.createIfMissing=false] - If true, creates the file with '{}' if it doesn't exist.
+ *        Set to true for write operations. Leave false for read-only operations that should handle
+ *        file-not-found scenarios in the callback.
+ * @returns {Promise<*>} Result of the callback
+ */
+async function withFileLock(filepath, callback, options = {}) {
+	const { createIfMissing = false } = options;
+	const fsPromises = fs.promises;
+
+	// Ensure parent directory exists
+	const dir = path.dirname(filepath);
+	await fsPromises.mkdir(dir, { recursive: true });
+
+	// Only create the file if explicitly requested (for write operations)
+	if (createIfMissing) {
+		try {
+			// Use 'wx' flag for atomic create - fails if file exists (prevents race)
+			await fsPromises.writeFile(filepath, '{}', { flag: 'wx' });
+		} catch (err) {
+			// EEXIST is expected if another process created the file - that's fine
+			if (err.code !== 'EEXIST') {
+				throw err;
+			}
+		}
+	}
+
+	const lockPath = `${filepath}.lock`;
+	const { maxRetries, retryDelay, staleLockAge } = LOCK_CONFIG;
+
+	// Try to acquire lock with retries
+	let acquired = false;
+	for (let attempt = 0; attempt < maxRetries; attempt++) {
+		try {
+			// Try to create lock file exclusively first
+			const lockContent = JSON.stringify({
+				pid: process.pid,
+				timestamp: Date.now()
+			});
+			await fsPromises.writeFile(lockPath, lockContent, { flag: 'wx' });
+			acquired = true;
+			break;
+		} catch (err) {
+			if (err.code === 'EEXIST') {
+				// Lock file exists - check if it's stale
+				try {
+					const lockStat = await fsPromises.stat(lockPath);
+					const age = Date.now() - lockStat.mtimeMs;
+					if (age > staleLockAge) {
+						// Stale lock - use atomic rename to safely take ownership
+						// This prevents race where we delete another process's fresh lock
+						const stalePath = `${lockPath}.stale.${process.pid}.${Date.now()}`;
+						try {
+							await fsPromises.rename(lockPath, stalePath);
+							// We successfully took ownership of the stale lock
+							// Clean it up and retry immediately
+							try {
+								await fsPromises.unlink(stalePath);
+							} catch {
+								// Ignore cleanup errors
+							}
+							continue; // Retry lock acquisition
+						} catch {
+							// Rename failed - another process handled it or lock was refreshed
+							// Just continue to retry
+						}
+					}
+				} catch (statErr) {
+					// Lock was removed between writeFile and stat - retry immediately
+					if (statErr.code === 'ENOENT') {
+						continue;
+					}
+					throw statErr;
+				}
+
+				// Lock exists and isn't stale (or we couldn't handle it), wait and retry
+				if (attempt < maxRetries - 1) {
+					const waitMs = retryDelay * Math.pow(2, attempt);
+					await sleep(waitMs);
+				}
+			} else {
+				throw err;
+			}
+		}
+	}
+
+	if (!acquired) {
+		throw new Error(
+			`Failed to acquire lock on ${filepath} after ${maxRetries} attempts`
+		);
+	}
+
+	try {
+		return await callback();
+	} finally {
+		// Release lock
+		try {
+			await fsPromises.unlink(lockPath);
+		} catch (releaseError) {
+			// Always log lock release failures - they indicate potential issues
+			log(
+				'warn',
+				`Failed to release lock for ${filepath}: ${releaseError.message}`
+			);
+		}
+	}
+}
+
+/**
+ * Synchronous version of file locking for compatibility with existing sync code
+ * Uses a lock file approach with retries and stale lock detection
+ * @param {string} filepath - Path to the file to lock
+ * @param {Function} callback - Sync function to execute while holding the lock
+ * @param {Object} [options] - Options for lock behavior
+ * @param {boolean} [options.createIfMissing=false] - If true, creates the file with '{}' if it doesn't exist.
+ *        Set to true for write operations. Leave false for read-only operations that should handle
+ *        file-not-found scenarios in the callback.
+ * @returns {*} Result of the callback
+ */
+function withFileLockSync(filepath, callback, options = {}) {
+	const { createIfMissing = false } = options;
+
+	// Ensure parent directory exists
+	const dir = path.dirname(filepath);
+	if (!fs.existsSync(dir)) {
+		fs.mkdirSync(dir, { recursive: true });
+	}
+
+	// Only create the file if explicitly requested (for write operations)
+	if (createIfMissing) {
+		try {
+			// Use 'wx' flag for atomic create - fails if file exists (prevents race)
+			fs.writeFileSync(filepath, '{}', { flag: 'wx' });
+		} catch (err) {
+			// EEXIST is expected if another process created the file - that's fine
+			if (err.code !== 'EEXIST') {
+				throw err;
+			}
+		}
+	}
+
+	const lockPath = `${filepath}.lock`;
+	const { maxRetries, retryDelay, staleLockAge } = LOCK_CONFIG;
+
+	// Try to acquire lock with retries
+	let acquired = false;
+	for (let attempt = 0; attempt < maxRetries; attempt++) {
+		try {
+			// Try to create lock file exclusively first
+			const lockContent = JSON.stringify({
+				pid: process.pid,
+				timestamp: Date.now()
+			});
+			fs.writeFileSync(lockPath, lockContent, { flag: 'wx' });
+			acquired = true;
+			break;
+		} catch (err) {
+			if (err.code === 'EEXIST') {
+				// Lock file exists - check if it's stale
+				try {
+					const lockStat = fs.statSync(lockPath);
+					const age = Date.now() - lockStat.mtimeMs;
+					if (age > staleLockAge) {
+						// Stale lock - use atomic rename to safely take ownership
+						// This prevents race where we delete another process's fresh lock
+						const stalePath = `${lockPath}.stale.${process.pid}.${Date.now()}`;
+						try {
+							fs.renameSync(lockPath, stalePath);
+							// We successfully took ownership of the stale lock
+							// Clean it up and retry immediately
+							try {
+								fs.unlinkSync(stalePath);
+							} catch {
+								// Ignore cleanup errors
+							}
+							continue; // Retry lock acquisition
+						} catch {
+							// Rename failed - another process handled it or lock was refreshed
+							// Just continue to retry
+						}
+					}
+				} catch (statErr) {
+					// Lock was removed between writeFile and stat - retry immediately
+					if (statErr.code === 'ENOENT') {
+						continue;
+					}
+					throw statErr;
+				}
+
+				// Lock exists and isn't stale (or we couldn't handle it), wait and retry
+				if (attempt < maxRetries - 1) {
+					const waitMs = retryDelay * Math.pow(2, attempt);
+					sleepSync(waitMs);
+				}
+			} else {
+				throw err;
+			}
+		}
+	}
+
+	if (!acquired) {
+		throw new Error(
+			`Failed to acquire lock on ${filepath} after ${maxRetries} attempts`
+		);
+	}
+
+	try {
+		return callback();
+	} finally {
+		// Release lock
+		try {
+			fs.unlinkSync(lockPath);
+		} catch (releaseError) {
+			// Always log lock release failures - they indicate potential issues
+			log(
+				'warn',
+				`Failed to release lock for ${filepath}: ${releaseError.message}`
+			);
+		}
+	}
+}
+
 // --- Environment Variable Resolution Utility ---
 /**
  * Resolves an environment variable's value.
@@ -703,6 +975,7 @@ function markMigrationForNotice(tasksJsonPath) {
 
 /**
  * Writes and saves a JSON file. Handles tagged task lists properly.
+ * Uses cross-process file locking and atomic writes to prevent race conditions.
  * @param {string} filepath - Path to the JSON file
  * @param {Object} data - Data to write (can be resolved tag data or raw tagged data)
  * @param {string} projectRoot - Optional project root for tag context
@@ -712,107 +985,172 @@ function writeJSON(filepath, data, projectRoot = null, tag = null) {
 	const isDebug = process.env.TASKMASTER_DEBUG === 'true';
 
 	try {
-		let finalData = data;
+		// Use file locking to prevent concurrent write race conditions
+		// This ensures the entire read-modify-write cycle is atomic
+		// createIfMissing: true because writeJSON is a write operation
+		withFileLockSync(
+			filepath,
+			() => {
+				let finalData = data;
 
-		// If data represents resolved tag data but lost _rawTaggedData (edge-case observed in MCP path)
-		if (
-			!data._rawTaggedData &&
-			projectRoot &&
-			Array.isArray(data.tasks) &&
-			!hasTaggedStructure(data)
-		) {
-			const resolvedTag = tag || getCurrentTag(projectRoot);
+				// If data represents resolved tag data but lost _rawTaggedData (edge-case observed in MCP path)
+				if (
+					data &&
+					!data._rawTaggedData &&
+					projectRoot &&
+					Array.isArray(data.tasks) &&
+					!hasTaggedStructure(data)
+				) {
+					const resolvedTag = tag || getCurrentTag(projectRoot);
 
-			if (isDebug) {
-				console.log(
-					`writeJSON: Detected resolved tag data missing _rawTaggedData. Re-reading raw data to prevent data loss for tag '${resolvedTag}'.`
-				);
-			}
+					if (isDebug) {
+						console.log(
+							`writeJSON: Detected resolved tag data missing _rawTaggedData. Re-reading raw data to prevent data loss for tag '${resolvedTag}'.`
+						);
+					}
 
-			// Re-read the full file to get the complete tagged structure
-			const rawFullData = JSON.parse(fs.readFileSync(filepath, 'utf8'));
-
-			// Merge the updated data into the full structure
-			finalData = {
-				...rawFullData,
-				[resolvedTag]: {
-					// Preserve existing tag metadata if it exists, otherwise use what's passed
-					...(rawFullData[resolvedTag]?.metadata || {}),
-					...(data.metadata ? { metadata: data.metadata } : {}),
-					tasks: data.tasks // The updated tasks array is the source of truth here
-				}
-			};
-		}
-		// If we have _rawTaggedData, this means we're working with resolved tag data
-		// and need to merge it back into the full tagged structure
-		else if (data && data._rawTaggedData && projectRoot) {
-			const resolvedTag = tag || getCurrentTag(projectRoot);
-
-			// Get the original tagged data
-			const originalTaggedData = data._rawTaggedData;
-
-			// Create a clean copy of the current resolved data (without internal properties)
-			const { _rawTaggedData, tag: _, ...cleanResolvedData } = data;
-
-			// Update the specific tag with the resolved data
-			finalData = {
-				...originalTaggedData,
-				[resolvedTag]: cleanResolvedData
-			};
-
-			if (isDebug) {
-				console.log(
-					`writeJSON: Merging resolved data back into tag '${resolvedTag}'`
-				);
-			}
-		}
-
-		// Clean up any internal properties that shouldn't be persisted
-		let cleanData = finalData;
-		if (cleanData && typeof cleanData === 'object') {
-			// Remove any _rawTaggedData or tag properties from root level
-			const { _rawTaggedData, tag: tagProp, ...rootCleanData } = cleanData;
-			cleanData = rootCleanData;
-
-			// Additional cleanup for tag objects
-			if (typeof cleanData === 'object' && !Array.isArray(cleanData)) {
-				const finalCleanData = {};
-				for (const [key, value] of Object.entries(cleanData)) {
-					if (
-						value &&
-						typeof value === 'object' &&
-						Array.isArray(value.tasks)
-					) {
-						// This is a tag object - clean up any rogue root-level properties
-						const { created, description, ...cleanTagData } = value;
-
-						// Only keep the description if there's no metadata.description
-						if (
-							description &&
-							(!cleanTagData.metadata || !cleanTagData.metadata.description)
-						) {
-							cleanTagData.description = description;
+					// Re-read the full file to get the complete tagged structure
+					// This is now safe because we hold the lock
+					let rawFullData = {};
+					try {
+						rawFullData = JSON.parse(fs.readFileSync(filepath, 'utf8'));
+					} catch (readError) {
+						// File might be empty or invalid, start fresh
+						if (isDebug) {
+							console.log(
+								`writeJSON: Could not read existing file, starting fresh: ${readError.message}`
+							);
 						}
+					}
 
-						finalCleanData[key] = cleanTagData;
-					} else {
-						finalCleanData[key] = value;
+					// Merge the updated data into the full structure
+					finalData = {
+						...rawFullData,
+						[resolvedTag]: {
+							// Preserve existing tag metadata, merged with any new metadata
+							metadata: {
+								...(rawFullData[resolvedTag]?.metadata || {}),
+								...(data.metadata || {})
+							},
+							tasks: data.tasks // The updated tasks array is the source of truth here
+						}
+					};
+				}
+				// If we have _rawTaggedData, this means we're working with resolved tag data
+				// and need to merge it back into the full tagged structure
+				else if (data && data._rawTaggedData && projectRoot) {
+					const resolvedTag = tag || getCurrentTag(projectRoot);
+
+					// IMPORTANT: Re-read the file to get the CURRENT state instead of using
+					// potentially stale _rawTaggedData. This prevents lost updates from other processes.
+					let currentTaggedData;
+					try {
+						currentTaggedData = JSON.parse(fs.readFileSync(filepath, 'utf8'));
+					} catch (readError) {
+						// Fall back to _rawTaggedData if file can't be read
+						currentTaggedData = data._rawTaggedData;
+						if (isDebug) {
+							console.log(
+								`writeJSON: Using _rawTaggedData as fallback: ${readError.message}`
+							);
+						}
+					}
+
+					// Create a clean copy of the current resolved data (without internal properties)
+					const { _rawTaggedData, tag: _, ...cleanResolvedData } = data;
+
+					// Update the specific tag with the resolved data, preserving other tags
+					finalData = {
+						...currentTaggedData,
+						[resolvedTag]: cleanResolvedData
+					};
+
+					if (isDebug) {
+						console.log(
+							`writeJSON: Merging resolved data back into tag '${resolvedTag}'`
+						);
 					}
 				}
-				cleanData = finalCleanData;
-			}
-		}
 
-		fs.writeFileSync(filepath, JSON.stringify(cleanData, null, 2), 'utf8');
+				// Clean up any internal properties that shouldn't be persisted
+				let cleanData = finalData;
+				if (cleanData && typeof cleanData === 'object') {
+					// Remove any _rawTaggedData or tag properties from root level
+					const { _rawTaggedData, tag: tagProp, ...rootCleanData } = cleanData;
+					cleanData = rootCleanData;
 
-		if (isDebug) {
-			console.log(`writeJSON: Successfully wrote to ${filepath}`);
-		}
+					// Additional cleanup for tag objects
+					if (typeof cleanData === 'object' && !Array.isArray(cleanData)) {
+						const finalCleanData = {};
+						for (const [key, value] of Object.entries(cleanData)) {
+							if (
+								value &&
+								typeof value === 'object' &&
+								Array.isArray(value.tasks)
+							) {
+								// This is a tag object - clean up any rogue root-level properties
+								// Move created/description to metadata if they're at root level
+								const { created, description, ...cleanTagData } = value;
+
+								// Ensure metadata object exists
+								if (!cleanTagData.metadata) {
+									cleanTagData.metadata = {};
+								}
+
+								// Preserve created timestamp in metadata if it exists at root level
+								if (created && !cleanTagData.metadata.created) {
+									cleanTagData.metadata.created = created;
+								}
+
+								// Preserve description in metadata if it exists at root level
+								if (description && !cleanTagData.metadata.description) {
+									cleanTagData.metadata.description = description;
+								}
+
+								finalCleanData[key] = cleanTagData;
+							} else {
+								finalCleanData[key] = value;
+							}
+						}
+						cleanData = finalCleanData;
+					}
+				}
+
+				// Use atomic write: write to temp file then rename
+				// This prevents partial writes from corrupting the file
+				const tempPath = `${filepath}.tmp.${process.pid}`;
+				try {
+					fs.writeFileSync(
+						tempPath,
+						JSON.stringify(cleanData, null, 2),
+						'utf8'
+					);
+					fs.renameSync(tempPath, filepath);
+				} catch (writeError) {
+					// Clean up temp file on failure
+					try {
+						if (fs.existsSync(tempPath)) {
+							fs.unlinkSync(tempPath);
+						}
+					} catch {
+						// Ignore cleanup errors
+					}
+					throw writeError;
+				}
+
+				if (isDebug) {
+					console.log(`writeJSON: Successfully wrote to ${filepath}`);
+				}
+			},
+			{ createIfMissing: true }
+		);
 	} catch (error) {
 		log('error', `Error writing JSON file ${filepath}:`, error.message);
 		if (isDebug) {
 			log('error', 'Full error details:', error);
 		}
+		// Re-throw so callers know the write failed
+		throw error;
 	}
 }
 
@@ -1616,5 +1954,7 @@ export {
 	flattenTasksWithSubtasks,
 	ensureTagMetadata,
 	stripAnsiCodes,
-	normalizeTaskIds
+	normalizeTaskIds,
+	withFileLock,
+	withFileLockSync
 };

tests/unit/file-locking.test.js

@@ -0,0 +1,595 @@
+/**
+ * Tests for file locking and atomic write functionality
+ * Verifies that concurrent access to tasks.json is properly serialized
+ */
+
+import fs from 'fs';
+import os from 'os';
+import path from 'path';
+import { fileURLToPath } from 'url';
+import { afterEach, beforeEach, describe, expect, it } from '@jest/globals';
+
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = path.dirname(__filename);
+
+// Import the utils module
+const utilsPath = path.join(__dirname, '../../scripts/modules/utils.js');
+
+describe('File Locking and Atomic Writes', () => {
+	let tempDir;
+	let testFilePath;
+	let utils;
+
+	beforeEach(async () => {
+		// Create a temp directory for each test
+		tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'taskmaster-test-'));
+		testFilePath = path.join(tempDir, 'tasks.json');
+
+		// Initialize with empty tasks structure
+		fs.writeFileSync(
+			testFilePath,
+			JSON.stringify(
+				{
+					master: {
+						tasks: [],
+						metadata: { created: new Date().toISOString() }
+					}
+				},
+				null,
+				2
+			)
+		);
+
+		// Import utils fresh for each test
+		utils = await import(utilsPath + `?cachebust=${Date.now()}`);
+	});
+
+	afterEach(() => {
+		// Clean up temp directory and any lock files
+		if (tempDir && fs.existsSync(tempDir)) {
+			fs.rmSync(tempDir, { recursive: true, force: true });
+		}
+	});
+
+	describe('withFileLockSync', () => {
+		it('should execute callback while holding lock', () => {
+			const result = utils.withFileLockSync(testFilePath, () => {
+				return 'callback executed';
+			});
+
+			expect(result).toBe('callback executed');
+		});
+
+		it('should release lock after callback completes', () => {
+			utils.withFileLockSync(testFilePath, () => {
+				// First lock
+			});
+
+			// Should be able to acquire lock again
+			const result = utils.withFileLockSync(testFilePath, () => {
+				return 'second lock acquired';
+			});
+
+			expect(result).toBe('second lock acquired');
+		});
+
+		it('should release lock even if callback throws', () => {
+			expect(() => {
+				utils.withFileLockSync(testFilePath, () => {
+					throw new Error('Test error');
+				});
+			}).toThrow('Test error');
+
+			// Should still be able to acquire lock
+			const result = utils.withFileLockSync(testFilePath, () => 'recovered');
+			expect(result).toBe('recovered');
+		});
+
+		it('should create file if createIfMissing is true', () => {
+			const newFilePath = path.join(tempDir, 'new-file.json');
+
+			utils.withFileLockSync(
+				newFilePath,
+				() => {
+					// Lock acquired on new file
+				},
+				{ createIfMissing: true }
+			);
+
+			expect(fs.existsSync(newFilePath)).toBe(true);
+		});
+
+		it('should not create file if createIfMissing is false (default)', () => {
+			const newFilePath = path.join(tempDir, 'should-not-exist.json');
+
+			utils.withFileLockSync(newFilePath, () => {
+				// Lock acquired, but file should not be created
+			});
+
+			expect(fs.existsSync(newFilePath)).toBe(false);
+		});
+
+		it('should clean up lock file after completion', () => {
+			utils.withFileLockSync(testFilePath, () => {
+				// Do something
+			});
+
+			// Lock file should be cleaned up
+			expect(fs.existsSync(`${testFilePath}.lock`)).toBe(false);
+		});
+
+		it('should clean up lock file even on error', () => {
+			try {
+				utils.withFileLockSync(testFilePath, () => {
+					throw new Error('Test error');
+				});
+			} catch {
+				// Expected
+			}
+
+			// Lock file should be cleaned up
+			expect(fs.existsSync(`${testFilePath}.lock`)).toBe(false);
+		});
+	});
+
+	describe('withFileLock (async)', () => {
+		it('should execute async callback while holding lock', async () => {
+			const result = await utils.withFileLock(testFilePath, async () => {
+				await new Promise((resolve) => setTimeout(resolve, 10));
+				return 'async callback executed';
+			});
+
+			expect(result).toBe('async callback executed');
+		});
+
+		it('should release lock after async callback completes', async () => {
+			await utils.withFileLock(testFilePath, async () => {
+				// First lock
+			});
+
+			// Should be able to acquire lock again
+			const result = await utils.withFileLock(testFilePath, async () => {
+				return 'second lock acquired';
+			});
+
+			expect(result).toBe('second lock acquired');
+		});
+
+		it('should release lock even if async callback rejects', async () => {
+			await expect(
+				utils.withFileLock(testFilePath, async () => {
+					throw new Error('Async error');
+				})
+			).rejects.toThrow('Async error');
+
+			// Should still be able to acquire lock
+			const result = await utils.withFileLock(
+				testFilePath,
+				async () => 'recovered'
+			);
+			expect(result).toBe('recovered');
+		});
+
+		it('should create file if createIfMissing is true', async () => {
+			const newFilePath = path.join(tempDir, 'new-async-file.json');
+
+			await utils.withFileLock(
+				newFilePath,
+				async () => {
+					// Lock acquired on new file
+				},
+				{ createIfMissing: true }
+			);
+
+			expect(fs.existsSync(newFilePath)).toBe(true);
+		});
+
+		it('should not create file if createIfMissing is false (default)', async () => {
+			const newFilePath = path.join(tempDir, 'should-not-exist-async.json');
+
+			await utils.withFileLock(newFilePath, async () => {
+				// Lock acquired, but file should not be created
+			});
+
+			expect(fs.existsSync(newFilePath)).toBe(false);
+		});
+
+		it('should clean up lock file after completion', async () => {
+			await utils.withFileLock(testFilePath, async () => {
+				// Do something
+			});
+
+			// Lock file should be cleaned up
+			expect(fs.existsSync(`${testFilePath}.lock`)).toBe(false);
+		});
+
+		it('should clean up lock file even on error', async () => {
+			try {
+				await utils.withFileLock(testFilePath, async () => {
+					throw new Error('Test error');
+				});
+			} catch {
+				// Expected
+			}
+
+			// Lock file should be cleaned up
+			expect(fs.existsSync(`${testFilePath}.lock`)).toBe(false);
+		});
+
+		it('should serialize truly concurrent writes', async () => {
+			const numConcurrentWrites = 5;
+			const writes = [];
+
+			for (let i = 0; i < numConcurrentWrites; i++) {
+				writes.push(
+					utils.withFileLock(testFilePath, async () => {
+						const data = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+						data.master.tasks.push({
+							id: String(data.master.tasks.length + 1)
+						});
+						fs.writeFileSync(testFilePath, JSON.stringify(data, null, 2));
+					})
+				);
+			}
+
+			await Promise.all(writes);
+
+			const finalData = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+			expect(finalData.master.tasks).toHaveLength(numConcurrentWrites);
+		});
+	});
+
+	describe('writeJSON atomic writes', () => {
+		it('should not leave temp files on success', () => {
+			// Create a tagged structure that writeJSON expects
+			const taggedData = {
+				master: {
+					tasks: [{ id: '1', title: 'Test task', status: 'pending' }],
+					metadata: { created: new Date().toISOString() }
+				}
+			};
+
+			utils.writeJSON(testFilePath, taggedData, null, null);
+
+			const files = fs.readdirSync(tempDir);
+			const tempFiles = files.filter((f) => f.includes('.tmp'));
+			expect(tempFiles).toHaveLength(0);
+		});
+
+		it('should preserve data from other tags when writing to one tag', () => {
+			// Set up initial data with multiple tags
+			const initialData = {
+				master: {
+					tasks: [{ id: '1', title: 'Master task', status: 'pending' }],
+					metadata: { created: new Date().toISOString() }
+				},
+				feature: {
+					tasks: [{ id: '1', title: 'Feature task', status: 'pending' }],
+					metadata: { created: new Date().toISOString() }
+				}
+			};
+			fs.writeFileSync(testFilePath, JSON.stringify(initialData, null, 2));
+
+			// Write directly with tagged structure (simulating what commands do internally)
+			const updatedData = {
+				...initialData,
+				master: {
+					...initialData.master,
+					tasks: [
+						{ id: '1', title: 'Updated master task', status: 'pending' },
+						{ id: '2', title: 'New task', status: 'pending' }
+					]
+				}
+			};
+
+			utils.writeJSON(testFilePath, updatedData, null, null);
+
+			const written = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+
+			// Master should be updated
+			expect(written.master.tasks).toHaveLength(2);
+			expect(written.master.tasks[0].title).toBe('Updated master task');
+
+			// Feature should be preserved
+			expect(written.feature.tasks).toHaveLength(1);
+			expect(written.feature.tasks[0].title).toBe('Feature task');
+		});
+
+		it('should not leave lock files on success', () => {
+			const taggedData = {
+				master: {
+					tasks: [{ id: '1', title: 'Test task', status: 'pending' }],
+					metadata: {}
+				}
+			};
+
+			utils.writeJSON(testFilePath, taggedData, null, null);
+
+			expect(fs.existsSync(`${testFilePath}.lock`)).toBe(false);
+		});
+	});
+
+	describe('Concurrent write simulation', () => {
+		it('should handle rapid sequential writes without data loss', () => {
+			// Perform many rapid writes
+			const numWrites = 10;
+
+			for (let i = 0; i < numWrites; i++) {
+				// Read current data
+				let currentData;
+				try {
+					currentData = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+				} catch {
+					currentData = { master: { tasks: [], metadata: {} } };
+				}
+
+				// Add a new task
+				currentData.master.tasks.push({
+					id: String(i + 1),
+					title: `Task ${i + 1}`,
+					status: 'pending'
+				});
+
+				// Write with locking
+				utils.writeJSON(testFilePath, currentData, null, null);
+			}
+
+			const finalData = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+			expect(finalData.master.tasks).toHaveLength(numWrites);
+		});
+	});
+
+	describe('True concurrent process writes', () => {
+		it('should handle multiple processes writing simultaneously without data loss', async () => {
+			const { spawn } = await import('child_process');
+
+			const numProcesses = 5;
+			const tasksPerProcess = 3;
+
+			// Create a worker script with inline locking implementation
+			// This mirrors the withFileLockSync implementation but without external dependencies
+			const workerScript = `
+				import fs from 'fs';
+
+				const filepath = process.argv[2];
+				const processId = process.argv[3];
+				const numTasks = parseInt(process.argv[4], 10);
+
+				const LOCK_CONFIG = {
+					maxRetries: 10,
+					retryDelay: 50,
+					staleLockAge: 10000
+				};
+
+				function sleepSync(ms) {
+					const end = Date.now() + ms;
+					while (Date.now() < end) {
+						// Busy wait
+					}
+				}
+
+				function withFileLockSync(filepath, callback) {
+					const lockPath = filepath + '.lock';
+					const { maxRetries, retryDelay, staleLockAge } = LOCK_CONFIG;
+
+					let acquired = false;
+					for (let attempt = 0; attempt < maxRetries; attempt++) {
+						try {
+							const lockContent = JSON.stringify({
+								pid: process.pid,
+								timestamp: Date.now()
+							});
+							fs.writeFileSync(lockPath, lockContent, { flag: 'wx' });
+							acquired = true;
+							break;
+						} catch (err) {
+							if (err.code === 'EEXIST') {
+								try {
+									const lockStat = fs.statSync(lockPath);
+									const age = Date.now() - lockStat.mtimeMs;
+									if (age > staleLockAge) {
+										const stalePath = lockPath + '.stale.' + process.pid + '.' + Date.now();
+										try {
+											fs.renameSync(lockPath, stalePath);
+											try { fs.unlinkSync(stalePath); } catch {}
+											continue;
+										} catch {}
+									}
+								} catch (statErr) {
+									if (statErr.code === 'ENOENT') continue;
+									throw statErr;
+								}
+								if (attempt < maxRetries - 1) {
+									const waitMs = retryDelay * Math.pow(2, attempt);
+									sleepSync(waitMs);
+								}
+							} else {
+								throw err;
+							}
+						}
+					}
+
+					if (!acquired) {
+						throw new Error('Failed to acquire lock on ' + filepath + ' after ' + maxRetries + ' attempts');
+					}
+
+					try {
+						return callback();
+					} finally {
+						try {
+							fs.unlinkSync(lockPath);
+						} catch {}
+					}
+				}
+
+				async function main() {
+					for (let i = 0; i < numTasks; i++) {
+						withFileLockSync(filepath, () => {
+							let currentData;
+							try {
+								currentData = JSON.parse(fs.readFileSync(filepath, 'utf8'));
+							} catch {
+								currentData = { master: { tasks: [], metadata: {} } };
+							}
+
+							currentData.master.tasks.push({
+								id: 'P' + processId + '-' + (i + 1),
+								title: 'Task from process ' + processId + ' #' + (i + 1),
+								status: 'pending'
+							});
+
+							fs.writeFileSync(filepath, JSON.stringify(currentData, null, 2), 'utf8');
+						});
+
+						// Small delay to increase chance of interleaving
+						await new Promise(r => setTimeout(r, 10));
+					}
+				}
+
+				main().catch(err => {
+					console.error(err);
+					process.exit(1);
+				});
+			`;
+
+			// Write worker script to temp file
+			const workerPath = path.join(tempDir, 'worker.mjs');
+			fs.writeFileSync(workerPath, workerScript);
+
+			// Spawn multiple processes that write concurrently
+			const processes = [];
+			for (let i = 0; i < numProcesses; i++) {
+				const proc = spawn(
+					'node',
+					[workerPath, testFilePath, String(i), String(tasksPerProcess)],
+					{
+						stdio: ['ignore', 'pipe', 'pipe']
+					}
+				);
+				processes.push(
+					new Promise((resolve, reject) => {
+						let stderr = '';
+						proc.stderr.on('data', (data) => {
+							stderr += data.toString();
+						});
+						proc.on('close', (code) => {
+							if (code === 0) {
+								resolve();
+							} else {
+								reject(
+									new Error(`Process ${i} exited with code ${code}: ${stderr}`)
+								);
+							}
+						});
+						proc.on('error', reject);
+					})
+				);
+			}
+
+			// Wait for all processes to complete
+			await Promise.all(processes);
+
+			// Verify all tasks were written
+			const finalData = JSON.parse(fs.readFileSync(testFilePath, 'utf8'));
+			const expectedTasks = numProcesses * tasksPerProcess;
+
+			expect(finalData.master.tasks.length).toBe(expectedTasks);
+
+			// Verify we have tasks from all processes
+			for (let i = 0; i < numProcesses; i++) {
+				const tasksFromProcess = finalData.master.tasks.filter((t) =>
+					t.id.startsWith(`P${i}-`)
+				);
+				expect(tasksFromProcess.length).toBe(tasksPerProcess);
+			}
+		}, 30000); // 30 second timeout for concurrent test
+	});
+});
+
+describe('readJSON', () => {
+	let tempDir;
+	let testFilePath;
+	let utils;
+
+	beforeEach(async () => {
+		tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'taskmaster-test-'));
+		testFilePath = path.join(tempDir, 'tasks.json');
+
+		// Create .taskmaster directory for state.json
+		fs.mkdirSync(path.join(tempDir, '.taskmaster'), { recursive: true });
+		fs.writeFileSync(
+			path.join(tempDir, '.taskmaster', 'state.json'),
+			JSON.stringify({
+				currentTag: 'master'
+			})
+		);
+
+		utils = await import(utilsPath + `?cachebust=${Date.now()}`);
+	});
+
+	afterEach(() => {
+		if (tempDir && fs.existsSync(tempDir)) {
+			fs.rmSync(tempDir, { recursive: true, force: true });
+		}
+	});
+
+	it('should read tagged task data correctly', () => {
+		const data = {
+			master: {
+				tasks: [{ id: '1', title: 'Test', status: 'pending' }],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(testFilePath, JSON.stringify(data, null, 2));
+
+		const result = utils.readJSON(testFilePath, tempDir, 'master');
+
+		expect(result.tasks).toHaveLength(1);
+		expect(result.tasks[0].title).toBe('Test');
+	});
+
+	it('should return null for non-existent file', () => {
+		const result = utils.readJSON(path.join(tempDir, 'nonexistent.json'));
+		expect(result).toBeNull();
+	});
+});
+
+describe('Lock file stale detection', () => {
+	let tempDir;
+	let testFilePath;
+	let utils;
+
+	beforeEach(async () => {
+		tempDir = fs.mkdtempSync(path.join(os.tmpdir(), 'taskmaster-test-'));
+		testFilePath = path.join(tempDir, 'tasks.json');
+		fs.writeFileSync(testFilePath, '{}');
+		utils = await import(utilsPath + `?cachebust=${Date.now()}`);
+	});
+
+	afterEach(() => {
+		if (tempDir && fs.existsSync(tempDir)) {
+			fs.rmSync(tempDir, { recursive: true, force: true });
+		}
+	});
+
+	it('should remove stale lock files', () => {
+		const lockPath = `${testFilePath}.lock`;
+
+		// Create a lock file with old timestamp
+		fs.writeFileSync(
+			lockPath,
+			JSON.stringify({
+				pid: 99999, // Non-existent PID
+				timestamp: Date.now() - 20000 // 20 seconds ago
+			})
+		);
+
+		// Touch the file to make it old
+		const pastTime = new Date(Date.now() - 20000);
+		fs.utimesSync(lockPath, pastTime, pastTime);
+
+		// Should be able to acquire lock despite existing lock file
+		const result = utils.withFileLockSync(testFilePath, () => 'acquired');
+		expect(result).toBe('acquired');
+	});
+});

tests/unit/task-manager/tag-management.test.js

@@ -113,3 +113,180 @@ describe('Tag Management – writeJSON context preservation', () => {
 		expect(tagNames).not.toContain('copy');
 	});
 });
+
+describe('Tag Management – ready tasks count', () => {
+	beforeEach(() => {
+		fs.mkdirSync(TEMP_DIR, { recursive: true });
+	});
+
+	afterEach(() => {
+		fs.rmSync(TEMP_DIR, { recursive: true, force: true });
+	});
+
+	it('should count tasks with no dependencies as ready', async () => {
+		const data = {
+			master: {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'pending', dependencies: [] },
+					{ id: 2, title: 'Task 2', status: 'pending', dependencies: [] },
+					{ id: 3, title: 'Task 3', status: 'done', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		const masterTag = result.tags.find((t) => t.name === 'master');
+		expect(masterTag.readyTasks).toBe(2); // 2 pending, 1 done (not ready)
+	});
+
+	it('should count tasks with satisfied dependencies as ready', async () => {
+		const data = {
+			master: {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'done', dependencies: [] },
+					{ id: 2, title: 'Task 2', status: 'pending', dependencies: [1] }, // deps satisfied
+					{ id: 3, title: 'Task 3', status: 'pending', dependencies: [2] } // deps NOT satisfied
+				],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		const masterTag = result.tags.find((t) => t.name === 'master');
+		expect(masterTag.readyTasks).toBe(1); // only task 2 is ready
+	});
+
+	it('should exclude deferred and blocked tasks from ready count', async () => {
+		const data = {
+			master: {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'pending', dependencies: [] },
+					{ id: 2, title: 'Task 2', status: 'deferred', dependencies: [] },
+					{ id: 3, title: 'Task 3', status: 'blocked', dependencies: [] },
+					{ id: 4, title: 'Task 4', status: 'in-progress', dependencies: [] },
+					{ id: 5, title: 'Task 5', status: 'review', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		const masterTag = result.tags.find((t) => t.name === 'master');
+		// Only pending, in-progress, review are actionable
+		expect(masterTag.readyTasks).toBe(3); // tasks 1, 4, 5
+	});
+});
+
+describe('Tag Management – --ready filter', () => {
+	beforeEach(() => {
+		fs.mkdirSync(TEMP_DIR, { recursive: true });
+	});
+
+	afterEach(() => {
+		fs.rmSync(TEMP_DIR, { recursive: true, force: true });
+	});
+
+	it('should filter out tags with no ready tasks when --ready is set', async () => {
+		const data = {
+			'has-ready': {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'pending', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			},
+			'no-ready': {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'done', dependencies: [] },
+					{ id: 2, title: 'Task 2', status: 'deferred', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			},
+			'all-blocked': {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'blocked', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true, ready: true },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		expect(result.tags.length).toBe(1);
+		expect(result.tags[0].name).toBe('has-ready');
+		expect(result.totalTags).toBe(1);
+	});
+
+	it('should include all tags when --ready is not set', async () => {
+		const data = {
+			'has-ready': {
+				tasks: [
+					{ id: 1, title: 'Task 1', status: 'pending', dependencies: [] }
+				],
+				metadata: { created: new Date().toISOString() }
+			},
+			'no-ready': {
+				tasks: [{ id: 1, title: 'Task 1', status: 'done', dependencies: [] }],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true, ready: false },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		expect(result.tags.length).toBe(2);
+		expect(result.totalTags).toBe(2);
+	});
+
+	it('should return empty list when no tags have ready tasks', async () => {
+		const data = {
+			'all-done': {
+				tasks: [{ id: 1, title: 'Task 1', status: 'done', dependencies: [] }],
+				metadata: { created: new Date().toISOString() }
+			}
+		};
+		fs.writeFileSync(TASKS_PATH, JSON.stringify(data, null, 2));
+
+		const result = await listTags(
+			TASKS_PATH,
+			{ showTaskCounts: true, ready: true },
+			{ projectRoot: TEMP_DIR },
+			'json'
+		);
+
+		expect(result.tags.length).toBe(0);
+		expect(result.totalTags).toBe(0);
+	});
+});

turbo.json

@@ -39,6 +39,36 @@
 				"!{packages,apps}/**/node_modules/**"
 			],
 			"outputLogs": "new-only"
+		},
+		"test": {
+			"dependsOn": ["^build"],
+			"inputs": [
+				"$TURBO_DEFAULT$",
+				"!{packages,apps}/**/dist/**",
+				"!{packages,apps}/**/node_modules/**"
+			],
+			"outputLogs": "new-only",
+			"cache": false
+		},
+		"test:unit": {
+			"dependsOn": ["^build"],
+			"inputs": [
+				"$TURBO_DEFAULT$",
+				"!{packages,apps}/**/dist/**",
+				"!{packages,apps}/**/node_modules/**"
+			],
+			"outputLogs": "new-only",
+			"cache": false
+		},
+		"test:integration": {
+			"dependsOn": ["^build"],
+			"inputs": [
+				"$TURBO_DEFAULT$",
+				"!{packages,apps}/**/dist/**",
+				"!{packages,apps}/**/node_modules/**"
+			],
+			"outputLogs": "new-only",
+			"cache": false
 		}
 	},
 	"globalDependencies": ["turbo.json", "tsconfig.json", ".env*"]