A tool for extracting relevant coding rules and guidelines from a documentation repository. Analyzes a collection of markdown files and returns contextually relevant information based on a given task.
Single tool 'coderules' has a well-structured description (194 chars) explaining WHAT, WHEN, and WHY, which exceeds baseline (avg 194 chars). Input schema is present with two required parameters (task, docsPath), both typed as strings with descriptions. However, the tool name lacks an action verb (should be 'extract_coderules' or 'search_coderules'), violating the verb_noun convention. Output schema is not documented, the code returns formatted markdown but LLMs cannot see the structure. No error handling guidance is provided; errors are caught but not categorized or actionable. Parameters lack constraints (e.g., docsPath format, task length bounds). The tool performs a single, focused action (good composition), but lacks the polish expected of production tools.
A tool for extracting relevant coding rules and guidelines from a documentation repository. This tool analyzes a collection of markdown files and returns contextually relevant information based on a given task. When to use this tool: - When implementing new features that need to follow project guidelines - When modifying existing code and need to understand the project's conventions - When reviewing code to ensure compliance with established standards - When onboarding to a new project and need to understand its coding standards Parameters explained: - task: Description of what you're trying to do (implement feature, fix bug, etc.) - docsPath: Path to the directory containing markdown documentation files
Tool name 'coderules' lacks action verb. Should be 'extract_coderules', 'search_coderules', or 'get_coderules' to signal intent and match LLM parsing conventions.
Output schema not documented. Code returns formatted markdown content, but LLMs cannot see the structure (fields, types, pagination). Document: {type: 'object', properties: {content: {type: 'array', items: {type: 'object', properties: {type, text}}}}}.
Error handling lacks recovery guidance. Errors are caught in processRequest() but returned as generic {isError: true, content: [...]}. Should categorize as retryable/user-fixable and suggest next steps (e.g., 'Invalid docsPath. Verify directory exists.').
Parameters lack constraints. 'task' and 'docsPath' have no length limits, format patterns, or validation rules. Should specify: task (1-500 chars), docsPath (valid filesystem path, must exist).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 43 | 2026-07-28+ | v2 |
No pagination or result limits documented. If docs are large, returned content could exceed context window. Should add limit parameter (default 20, max 100) and document max result size.