Back to directory

fallow-skills

Agent skills teaching an AI coding assistant to use fallow, a static codebase-intelligence layer for TypeScript and JavaScript that reports unused code, duplication, circular dependencies and complexity hotspots in milliseconds.

fallow-skills

fallow-skills packages agent skills for fallow, a codebase intelligence layer for TypeScript and JavaScript. Fallow's static layer builds a full module graph and reports quality issues, changed-code risk, cleanup opportunities, circular dependencies, code duplication, complexity hotspots, architecture boundary violations and design-system styling drift in milliseconds, something the project says linters and the type checker cannot do on their own because neither builds the whole module graph and so neither can see what nothing depends on. An optional paid product, Fallow Cloud, adds production coverage data so an agent can safely delete cold code and flag risky changes to hot-path code with evidence.

The skills teach an agent which fallow commands to run, which flags to use, how to interpret the output, and how to avoid common mistakes, working with any agent that supports the Agent Skills specification, including Claude Code, Cursor, OpenAI Codex, Windsurf, GitHub Copilot, Gemini CLI and Amp.

They install through the Agent Skills CLI, a Claude Code or OpenAI Codex plugin marketplace command, or an agent-specific CLI shortcut, and require fallow itself to be installed in the target project, either as a prebuilt global binary or run directly with a package-runner command.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.