Back to directory

repository-harness

A repository protocol and installer that gives a coding agent a compact entrypoint, a navigable map, durable plans, and mechanical validation, so intent and decisions stay in the repo instead of chat.

repository-harness

repository-harness turns a software repository into what it calls a legible, agent-ready workspace. It installs a small protocol and a safe updater on top of an existing repository, which stays the system of record: product documents, decisions, plans, code, tests, CI and runtime evidence all define the work, rather than a separate task database or orchestrator.

It targets ordinary failure modes of coding agents: intent that only exists in chat, a repository that does not identify its own authoritative documents, small changes that acquire unnecessary process, long changes that lose their decisions, completion claimed without behavior-level proof, and an agent inventing policy when a request leaves a choice open. It installs a compact AGENTS.md entrypoint, a workflow and documentation map, templates for plans, decisions and runbooks, and onboarding and proposal-audit skills, installed with a curl or PowerShell script that supports merge, override and dry-run modes.

More in Coding agents

See all 793
  • The City

    github.com

    A native macOS app that shows Claude Code runs as a city, with each project as a building, each team as a floor of robots, and every animation driven by a real event from the claude CLI's stream.

  • rig

    tryrig.ai

    A small, single-binary agent runtime that points at any OpenAI-compatible model, local or hosted, reads and edits a repository's code, runs its tests and remembers what it learned, with a loop implemented in a few hundred lines.

  • Jauvex

    jauvex.reindent.com

    An Electron desktop app that puts Claude Code, Codex and Grok coding-agent sessions side by side and lets you steer them by voice, with a small model handling fast routing decisions.