Gavel

gavel 👨‍⚖️

  Multi-model fusion for Claude Code.
  Ask Claude + Codex + agy in parallel, then judge their answers into one and act on it.

/gavel:fuse  asks three models the same thing: the Claude Code model you're running, OpenAI Codex, and agy. Codex and agy are read-only advisors; Claude is panelist #3, the judge, and the actor. The three answers are synthesized into one fused answer, which Claude then acts on — and only Claude ever writes to your workspace.

It runs the models through their local CLIs (codex, agy), reusing your existing logins. No API keys to wire up, no MCP servers, no background jobs.

Inspiration

Gavel is inspired by OpenRouter's Fusion beats Frontier: dispatch a prompt to a panel of models, then have a judge synthesize their answers into one response that beats any single frontier model. Gavel brings that pattern into Claude Code — Codex and agy answer as advisors, and their answers are judged and fused into one before it acts.

Benchmark chart from OpenRouter's "Fusion beats Frontier" announcement (© OpenRouter), included for reference and attribution.

Install

In Claude Code:

(Reload/restart if prompted.)

Local development — clone the repo and point the marketplace at the clone instead:

Setup

Reports whether codex and agy are installed, authenticated, and recent enough, and offers to install Codex if it's missing. Authentication:

- Codex — !codex login (install: npm install -g @openai/codex).
- agy — run !agy once to sign in (Google OAuth). Install the agy CLI from  (it isn't an npm package).

Gavel needs at least one advisor usable, but works best with both.

Commands

How advisors stay read-only

Only Claude modifies your workspace. The advisors are constrained differently because their CLIs differ:

- Codex runs in your project…