Agent Readiness

Agent Readiness runs two rule-based engines on a repository on your machine. agent-readiness-kit scores readiness for AI coding agents from 0 to 100 across 13 categories (agent instructions, workflow, testing, safety boundaries, and more) for Node.js, Python, Go, and Rust projects. agent-context-doctor checks agent instruction and configuration files (AGENTS.md, CLAUDE.md, Cursor rules, MCP config) for vague or placeholder guidance, contradictions, stale commands, risky permissions, hidden Unicode, and committed secrets, which it masks. Every report names both engine commits. The scanner is read-only and offline: it writes nothing, runs none of the repository's scripts, and makes no network requests. The AI client that runs it may send the results, or files it reads itself, to its model provider. Credential masking is pattern-based and covers the scanner's output only. Not to be confused with tools that score websites for agents: this audits a source repository. Install (stdio, Node.js 22.13+): npx -y @alipajand/agent-readiness-mcp@0.1.2 --root /path/to/projects Tools: audit_repository, audit_agent_context, get_audit_report. Official MCP Registry: io.github.alipajand/agent-readiness.