$ loading_
Scan repositories, generate conventions, and guide AI agents to code consistently.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "patchwork — Codebase Conventions for AI Agents" MCP server from askskill: Run: claude mcp add 'io-github-sainarayana-b-patchwork' -- npx -y patchwork-conventions
Scan this codebase, identify naming, folder structure, testing patterns, commit habits, and code style, then generate a CONVENTIONS.md grouped by “required”, “recommended”, and “examples”.
A clear CONVENTIONS.md summarizing project conventions with actionable examples.
Generate and expose convention tools for AI from the current repository, then tell me what conventions should be followed when adding a new API endpoint, test file, and configuration entry.
Project conventions the AI can reference, plus concrete implementation rules and suggestions for the new feature.
After scanning the repository, produce a concise onboarding guide for new team members: list the most important coding conventions, common anti-patterns, pre-commit checks, and mistakes AI coding assistants often make.
A concise convention guide for newcomers to quickly learn project rules and reduce collaboration drift.
Scan codebases for prioritized findings and export results in structured formats.
Perform precise text operations for AI agents, including diffs, patches, and regex tests.
Scan AI-generated code for injections, secrets, SSRF, and risky patterns.
Gives AI coding agents a structural map of your repository fast.
Scan code for TODO, FIXME, and XXX markers with prioritized outputs.
Review code diffs for bugs, security risks, and generate fixes.