c59bd8c9fc
Fix issue with kebab-case validator incorrectly flagging single-word flags like --prompt. Refactor detectCamelCaseFlags to properly handle all single-word flags. Update tests to verify correct behavior with single-word and camelCase flags. Add support for alternative flag formats in init command (e.g., -my_name). This fixes a bug where users couldn't use the --prompt flag directly and had to use -p instead.
57 lines
2.8 KiB
Plaintext
57 lines
2.8 KiB
Plaintext
# Task ID: 32
|
|
# Title: Implement 'learn' Command for Automatic Cursor Rule Generation
|
|
# Status: pending
|
|
# Dependencies: None
|
|
# Priority: high
|
|
# Description: Create a new 'learn' command that analyzes code changes and chat history to automatically generate or update Cursor rules in the .cursor/rules directory based on successful implementation patterns.
|
|
# Details:
|
|
Implement a new command in the task-master CLI that enables Cursor to learn from successful coding patterns:
|
|
|
|
1. Create a new module `commands/learn.js` that implements the command logic
|
|
2. Update `index.js` to register the new command
|
|
3. The command should:
|
|
- Accept an optional parameter for specifying which patterns to focus on
|
|
- Use git diff to extract code changes since the last commit
|
|
- Access the Cursor chat history if possible (investigate API or file storage location)
|
|
- Call Claude via ai-services.js with the following context:
|
|
* Code diffs
|
|
* Chat history excerpts showing challenges and solutions
|
|
* Existing rules from .cursor/rules if present
|
|
- Parse Claude's response to extract rule definitions
|
|
- Create or update .mdc files in the .cursor/rules directory
|
|
- Provide a summary of what was learned and which rules were updated
|
|
|
|
4. Create helper functions to:
|
|
- Extract relevant patterns from diffs
|
|
- Format the prompt for Claude to focus on identifying reusable patterns
|
|
- Parse Claude's response into valid rule definitions
|
|
- Handle rule conflicts or duplications
|
|
|
|
5. Ensure the command handles errors gracefully, especially if chat history is inaccessible
|
|
6. Add appropriate logging to show the learning process
|
|
7. Document the command in the README.md file
|
|
|
|
# Test Strategy:
|
|
1. Unit tests:
|
|
- Create tests for each helper function in isolation
|
|
- Mock git diff responses and chat history data
|
|
- Verify rule extraction logic works with different input patterns
|
|
- Test error handling for various failure scenarios
|
|
|
|
2. Integration tests:
|
|
- Test the command in a repository with actual code changes
|
|
- Verify it correctly generates .mdc files in the .cursor/rules directory
|
|
- Check that generated rules follow the correct format
|
|
- Verify the command correctly updates existing rules without losing custom modifications
|
|
|
|
3. Manual testing scenarios:
|
|
- Run the command after implementing a feature with specific patterns
|
|
- Verify the generated rules capture the intended patterns
|
|
- Test the command with and without existing rules
|
|
- Verify the command works when chat history is available and when it isn't
|
|
- Test with large diffs to ensure performance remains acceptable
|
|
|
|
4. Validation:
|
|
- After generating rules, use them in Cursor to verify they correctly guide future implementations
|
|
- Have multiple team members test the command to ensure consistent results
|