Generates structured AGENTS.md files to provide AI coding agents with precise context, build commands, and operational boundaries.
The wiki-agents-md skill automates the creation of AI-specific documentation that bridges the gap between human-readable READMEs and the technical context required by autonomous coding agents. By scanning your repository's configuration files (like package.json, pyproject.toml, or Cargo.toml), it extracts real-world build, test, and run commands. This skill improves agent accuracy and reliability by explicitly defining project structures, coding styles, and strict 'Do/Ask/Never' boundaries, ensuring AI tools like Claude Code operate within your project's specific constraints without overwriting existing documentation.
主要功能
01Pattern detection for coding styles and naming conventions
023 GitHub stars
03Strict safety guards to prevent overwriting existing AGENTS.md files
04Companion CLAUDE.md generation to redirect AI tools to specialized guidance
05Tailored documentation generation for root, source, and library directories
06Automated extraction of build and test commands from repository manifests
使用场景
01Standardizing AI-readiness across multiple services in a monorepo
02Defining clear operational guardrails for AI-assisted development
03Preparing a complex repository for autonomous coding agents