Hone AI

hone-ai

AI Coding Agent Orchestrator — Automatically implement features from requirements using AI agents.

Transform feature ideas into working code through autonomous development with human oversight.

Why

When working on long running tasks with agents, their context window fills up and the performance degrades. To mitigate this, hone-ai provides a solution starting each new iteration with a fresh context window just passing in a summary of the progress made so far on the specific PRD, repository architecture and gotchas, and any other relevant information.

Everything else is stripped away, leaving only the essential information needed for the next iteration. This approach helps maintain optimal performance and reduces the likelihood of context drift.

It's a surprisingly powerful process.

Quick Start

Install hone as a native Claude Code plugin:

To update:

That's it! You're ready to use hone.

Recommended Workflow: /hone:auto

The preferred way to use hone is to plan first, then hand the plan to /hone:auto — it runs the entire chain (prd → prd-to-tasks → run → review → fix loop) hands-free.

/hone:auto stops once, up front, to ask batched PRD clarifying questions (or skips them entirely if codebase analysis answers everything), then drives every transition mechanically: it generates the PRD, breaks it into tasks, runs the implementation loop on its own feature branch (hone/), audits the branch, and loops review→fix until clean (capped by --max-rounds, default 3).

hone implements the feature, runs tests, and commits changes automatically when the project is version-controlled (git by default; other VCS like jj/hg/sl are detected and used in their place). If .plans/ is ignored by VCS, hone skips committing those files but still commits code changes. /hone:auto…