Back to directory

Agent Skills

A validated, security-scanned registry of packaged skills that extend coding agents such as Claude Code, Cursor, Copilot, and Antigravity with new workflows.

Visit website (opens in a new tab)agent-skills.techleads.club
Agent Skills

Agent Skills packages reusable workflows and specialized knowledge as installable skill folders, each containing a SKILL.md instruction file plus optional templates and reference docs, following the structure packages/skills-catalog/skills/(category)/skill/. The project positions itself against marketplaces where a share of published skills have been found to contain vulnerabilities, and instead runs every skill through static analysis, immutable lockfiles and content hashing, and a security scan such as Snyk Agent Scan before publishing.

Its CLI adds sanitization, path isolation, symlink guards, an atomic lockfile, and an audit trail when installing a skill. It targets a tiered list of agents split into popular tools such as Claude Code and Cursor, rising tools such as Aider and Gemini CLI, and enterprise tools such as Amazon Q and Tabnine.

It is written in TypeScript.

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.