hoangnb24/repository-harness

Turn any repo into an agent-ready workspace for Claude Code, Codex, Cursor, and other coding agents.

What it solves

Coding agents often struggle with a lack of clear intent, missing authoritative documents, and a tendency to invent product policy when faced with ambiguity. repository-harness solves this by transforming a software repository into a structured, agent-ready workspace that emphasizes evidence-backed changes and explicit decision boundaries.

How it works

The tool installs a repository protocol and a safe updater into a target project. It establishes a core set of files—including a compact AGENTS.md entrypoint, a documentation map, and structures for product decisions and execution plans—to guide agents. It defines a specific workflow for requests: read-only requests are answered with evidence, bounded changes are implemented with proof, and complex changes require durable plans in docs/plans/active/.

Who it’s for

Developers who use AI coding agents and want to ensure their agents follow a repository's authoritative documents, avoid speculative policy invention, and maintain a durable record of decisions and progress.

Highlights

  • Agent-Ready Structure: Provides a compact entrypoint and navigable map for agents to understand the repository.
  • Safe Maintenance: Includes a versioned binary updater that uses three-way merges to preserve local edits and handle conflicts transactionally.
  • Explicit Decision Boundaries: Forces agents to stop before making material product changes when ambiguity exists.
  • Invariant Enforcement: Offers optional skills for encoding invariants and explicit-only onboarding of brownfield repositories.

Related

  • Project
    WILLOSCAR/research-units-pipeline-skills

    A framework for agent-assisted research that transforms research goals into verifiable, resumable runs with a full evidence trail and pinned pipeline contracts.

  • Project
    Ancienttwo/repo-harness

    A file-backed workflow and runtime for AI agents that replaces chat memory with repository-local artifacts to ensure continuity and consistency across sessions.

  • Project
    ruvnet/metaharness

    A factory for minting custom, repo-aware AI agent harnesses with their own CLI, skills, and governance, compatible with multiple agent hosts.

  • Project
    affaan-m/ECC

    An agent harness operating system that provides AI agents with a coordinated engineering system, including 68 agents and 286 skills to implement professional software engineering workflows.

  • Project
    Darwin-Agent/HarnessX

    HarnessX is a composable agent harness foundry that decouples model selection from behavior logic, allowing developers to evolve agent performance through automated harness optimization and RL fine-tuning.