Back to directory

Skills for agents

A collection of small, composable skills that add capabilities to a coding agent, including PR review and shipping workflows, visual planning and editing tools, and utilities for managing usage limits and context.

Skills for agents

Skills for agents is a set of small, composable skills installed into a coding agent with a single command. An experimental Factory group of skills turns feedback, product telemetry, errors and delivery signals into policy-gated software changes, including collecting reports, following an authorized pull request through checks, reviewing a pull-request queue, publishing changes under configured verification, finding recurring issues over a time window, and finding stalled work.

Other skills open and operate companion apps or web apps that use MCP tools, turn text plans into visual plans or diffs into visual recaps, open a running local app for visual editing, compare competing plans, keep working through ordinary ambiguity, track usage limits during long-running work, and check authoritative documentation before guessing.

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.