Back to directory

Foreman

Foreman keeps a project roadmap beside the code for Claude Code, Codex and Antigravity, and recommends the next task with its reasoning checked against the current state of the files before you start it.

Foreman

Foreman is meant to preserve continuity between coding sessions: it records requested work with its reason and boundaries, tracks status, blockers and work waiting for acceptance, and when asked "what's next", checks a candidate task's assumptions against the project's actual files before recommending it. Optional lessons let a later task learn from earlier ones without maintaining a separate knowledge base, and acceptance of finished work is always kept as a separate decision from a passing test run.

It is used through plain-language requests or through each assistant's own command style - slash commands in Claude Code, a named skill in Codex, or a slash command in Antigravity - and installs as one plugin shared across all three.

It is written in JavaScript.

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.