
by YawLabs
ctxlint checks AI agent context files and MCP server configs against your real codebase and flags drift before your agent acts on it. It runs as a CLI, CI step, pre-commit hook or MCP server.
PluginContact for pricing
Step 1: Run npx -y @yawlabs/ctxlint@latest in your project, or install it with npm install -D @yawlabs/ctxlint.
Step 2: Read the report of broken paths, wrong commands, stale context and contradictions.
Step 3: Add --mcp to include MCP config files, or --mcp-only to lint only those; --mcp-global adds user-level client configs.
Step 4: Run with --fix to rewrite broken paths using git history.
Step 5: Add a script such as ctxlint --strict to package.json so CI or a pre-commit hook runs it.
Step 6: To use it from an agent, run ctxlint as an MCP server in your client.
It covers CLAUDE.md, AGENTS.md, .cursorrules, Copilot instructions, Windsurf rules, GEMINI.md, .clinerules and files for several other tools, 16 AI tools in total.
Yes. With --mcp or --mcp-only it scans files like .mcp.json, .cursor/mcp.json and .vscode/mcp.json for schema, security, command and URL problems.
Yes. It works as a CLI or CI step and can output SARIF for GitHub Code Scanning.
No. You can run it directly with npx, or add it as a dev dependency or global package.
Requirements
Electron MCPElectron MCP is an MCP server from Yaw Labs that gives AI coding assistants structured knowledge about Electron: IPC scaffolding, security audits, build-error diagnosis and version migration. It is for developers building Electron apps with an assistant.
YawLabs Redis MCPAn MCP server from Yaw Labs for exploring and diagnosing Redis and Valkey instances from an MCP client. It is read-only by default and always enumerates keys with SCAN.
YawLabs AWS MCP@yawlabs/aws-mcp is a small MCP server that calls AWS through your local aws CLI, with built-in SSO re-login, generic resource CRUD and live docs lookup. It is presented as an alternative to AWS's own MCP server.
Akash NetworkAkash Network is a decentralized marketplace where independent providers bid to host your containers and GPU workloads. You describe what you need and pick a bid.