RAAV vs AGENTS.md / CLAUDE.md
A flat AGENTS.md or CLAUDE.md is also memory. RAAV keeps that file and adds the wedge: a review queue to review, swimlanes to watch, and an audit trail to supervise.
Teams reach for AGENTS.md or CLAUDE.md as agent memory. It works until the founder needs to see what changed, who edited what, and whether it was verified — none of which a flat file shows.
RAAV starts where AGENTS.md stops. Memory is table stakes — a flat AGENTS.md is also memory. RAAV's difference is the founder-agent loop is visible: swimlanes to watch, a review queue to review, an audit trail to supervise.
This is not RAAV against memory — RAAV keeps your AGENTS.md. It is a flat file against a visible collaboration loop. The wedge is not storage; it is review, watch, and supervise.
An AGENTS.md or CLAUDE.md is simple, version-controlled, and readable by every agent. It carries conventions, setup, and operating rules with zero infrastructure.
RAAV does not replace that. It generates and syncs the file so your agents keep reading it.
A file cannot separate confirmed facts from assumptions, show which agent is editing which files, or record whether the last fix passed verification. It changes silently and the founder never sees the diff.
That is exactly the gap RAAV fills with review, watch, and supervise.
Review: proposals and memory candidates gate durable truth. Watch: swimlanes and claims show active work. Supervise: the audit trail records what shipped and passed.
Give your agent a memory it can operate through — and give yourself the surfaces to review, watch, and supervise it.
RAAV does not replace your coding agent. It gives the agent a durable operating system for product truth, task ownership, and proof.
RAAV generates and syncs your AGENTS.md and CLAUDE.md.
Changes route through a founder review queue.
Swimlanes and claims make parallel work visible.
The audit trail records verification and risk.
Ask your agent to install RAAV. It generates the AGENTS.md and adds the review, watch, and supervise surfaces on top.
Install RAAV in this repo. Generate and sync our AGENTS.md and CLAUDE.md, then add a review queue, swimlanes, and audit trail on top of the rules file.RAAV is local-first and zero-LLM by default. It is not a coding agent — it makes your rules file into reviewable memory.
Most tools either write code or track human tickets. RAAV sits between the founder and the coding agents as shared product memory, coordination, and audit.
This page answers the honest question: if AGENTS.md is already memory, why RAAV?
No. It generates and syncs your AGENTS.md, but the wedge is visibility: a review queue to review changes, swimlanes to watch work, and an audit trail to supervise agents — none of which a flat file provides.
No. RAAV keeps your project-specific content and syncs the operating rules across agent files.
No. RAAV is the memory and audit layer around coding agents. It is zero-LLM by default.
The local tier is free. Hosted Solo and Team plans are on the waitlist and not for sale yet.