The allowed-tools pattern was missing :* suffix, causing permission check
failures when arguments were passed to the setup script via ```! block.
Fixes#16398
Fixes#12170
The ralph-wiggum slash commands had multi-line bash scripts in their
```! blocks. Claude Code's security check blocks commands with
newlines to prevent command injection, causing the error:
"Command contains newlines that could separate multiple commands"
Changes:
ralph-loop.md:
- Remove multi-line bash from code block
- Keep single-line call to setup script
- Keep scoped allowed-tools for security
cancel-ralph.md:
- Replace multi-line bash with step-by-step instructions
- Tighten allowed-tools to specific file paths
setup-ralph-loop.sh:
- Add completion promise display logic (moved from ralph-loop.md)
- Uses COMPLETION_PROMISE variable directly instead of reading from file
Suggestions must include all necessary changes. If a fix requires
additional work elsewhere (e.g., renaming a variable requires updating
usages), use the Claude Code prompt format instead of a partial suggestion.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Only post a summary comment when no issues are found. When issues
exist, post inline comments directly without a redundant summary block.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Change the default behavior of /code-review to output to the terminal
instead of posting a PR comment. Users can use the --comment flag to
explicitly post the review as a PR comment when desired.
This is more suitable for local development workflows where posting
comments to the PR is not always needed.
Remove the "Generated with Claude Code" footer and feedback CTA
from the code review comment template as it adds noise without
providing value after the first viewing.
Adds the hookify plugin to public marketplace. Enables users to create custom
hooks using simple markdown configuration files instead of editing JSON.
Key features:
- Define rules with regex patterns to warn/block operations
- Create rules from explicit instructions or conversation analysis
- Pattern-based matching for bash commands, file edits, prompts, stop events
- Enable/disable rules dynamically without editing code
- Conversation analyzer agent finds problematic behaviors
Changes from internal version:
- Removed non-functional SessionStart hook (not registered in hooks.json)
- Removed all sessionstart documentation and examples
- Fixed restart documentation to consistently state "no restart needed"
- Changed license from "Internal Anthropic use only" to "MIT License"
- Kept test blocks in core modules (useful for developers)
Plugin provides:
- 4 commands: /hookify, /hookify:list, /hookify:configure, /hookify:help
- 1 agent: conversation-analyzer
- 1 skill: writing-rules
- 4 hook types: PreToolUse, PostToolUse, Stop, UserPromptSubmit
- 4 example rules ready to use
All features functional and suitable for public use.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Migrates the ralph-wiggum plugin from internal marketplace to public marketplace.
Implements Geoffrey Huntley's "Ralph Wiggum" technique using Claude Code's Stop
hook mechanism for continuous iterative development loops.
Key features:
- Interactive self-referential AI loops in current session
- Stop hook intercepts exit and feeds same prompt back
- Iteration tracking and completion promise detection
- Max iterations safety limits
Changes:
- Remove all tmux dependencies and background execution mode
- Simplify to interactive-only mode using Stop hooks
- Add comprehensive error handling with clear messages
- Fix documentation to accurately describe Stop hook mechanism
- Add input validation for all command-line arguments
- Register plugin in public marketplace
Security fixes:
- Remove eval usage (command injection vulnerability)
- Add numeric validation before arithmetic operations
- Remove silent error suppression
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add plugin metadata configuration file with version, description, and author information for the frontend-design plugin.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add a new plugin that helps create distinctive, production-grade frontend interfaces with high design quality. The plugin includes a skill that generates creative, polished code avoiding generic AI aesthetics.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update session-start.sh to include explanatory insights alongside learning mode
- Add educational insight formatting with ★ Insight sections
- Update README.md to clarify differences from unshipped Learning output style
- Document that this plugin now combines both learning and explanatory functionality
- Address review feedback about incorporating explanatory-output-style features
Co-authored-by: Boris Cherny <bcherny@users.noreply.github.com>
Add interactive learning mode plugin that requests meaningful code contributions at decision points. Based on the unshipped Learning output style, this plugin engages users in active learning by having them write 5-10 lines of code for business logic, error handling, and design decisions.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added plugin metadata files for code-review and commit-commands plugins
to comply with Claude plugin structure requirements. These files were
identified as missing by the [claudelint](https://github.com/stbenjam/claudelint)
tool, which validates plugin structure and format according to the Claude
Code plugin conventions.
Add new code-review plugin that provides automated pull request reviews using multiple specialized agents with confidence-based scoring to filter false positives.
Key features:
- Multiple parallel agents for independent auditing (CLAUDE.md compliance, bug detection, historical context)
- Confidence-based scoring (0-100) with 80+ threshold to filter false positives
- Automatic skipping of closed, draft, or already-reviewed PRs
- Links directly to code with full SHA and line ranges
Updates:
- Add code-review plugin directory with command and README
- Update plugins/README.md to document the new plugin
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add plugins section to main README with link to detailed docs
- Create plugins/README.md with overview of all available plugins
- Add detailed READMEs for agent-sdk-dev, commit-commands, and feature-dev plugins
- Document all commands, agents, usage patterns, and workflows
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
The model field is no longer needed in the command frontmatter as model selection is now handled at a different level.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
- Move plugin.json to .claude-plugin directory to match pr-review-toolkit pattern
- Add author information: Ashwin Bhat (ashwin@anthropic.com)
- Update model field from sonnet-4.5 to sonnet in all agent/command definitions
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Add bundled plugins from claude-code-marketplace into a new plugins/ directory:
- agent-sdk-dev: Development kit for Claude Agent SDK with TypeScript/Python verifiers
- ireview-plugin: Comprehensive AI+Human PR review toolkit with 9 specialized agents
- commit-commands: Git commit workflow commands (commit, push, PR creation)
Created .claude-plugin/marketplace.json at repo root to register the bundled plugins.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
