Relay Reviewer — Development agent for Claude Code
One round of the multi-model relay.
How to install Relay Reviewer
Installs to ~/.claude/agents/watsonbkeel-multi-model-relay-relay-reviewer.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/watsonbkeel/multi-model-relay/HEAD/agents/relay-reviewer.md -o ~/.claude/agents/watsonbkeel-multi-model-relay-relay-reviewer.md Restart Claude Code, or start a new session, for it to be picked up.
What Relay Reviewer does
name: relay-reviewer description: One round of the multi-model relay. Reads the original, the previous draft and the previous notes, writes an improved draft plus structured notes into its own round directory, and returns a one-line JSON. Invoke with an explicit `model` so each round runs on a different model. tools: Read, Write, Glob, Grep effort: high maxTurns: 20
You are one seat in a multi-model relay. The prompt you receive is a `
Alternatives in Development
- Harvest Curator — Draft session learnings from harvest staging into lizard-ready adoption notes 1k ★
- Dashclaw Gate Runner — Runs the DashClaw verification gates (lint, full vitest suite, build, contract checks) and returns ONLY the fa 297 ★
- Chain Executor — Executes delegated chain steps with context isolation 186 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Anti Pattern Selector
Selects a varied subset of anti-patterns for the current kern run. Reads the manifest plus site-specific signa
Paf Reviewer
Runs one phone-a-friend relay or review command in the background and reports back a receipt plus the findings
TurboFlow 4.0.0 Release Notes — The Ruflo Migration
Release Date: March 7, 2026 Previous Version: v3.4.1 (February 2025) ---
Doc Builder
Use when the written draft must be poured into the official form file keeping its original formatting. Trigger
GPT Chat
Conducts a conversation with a GPT model (Codex) as a separate conversational partner—the thread persists betw
Model Corroborator
Stage 5.5 of /analyze-policy. When the original reform has no exact-shape external comparator, runs the closes
Related Skills
Orbit Prompt
Analyze and improve your prompt before sending — returns ORIGINAL PROMPT, ANALYSIS, IMPROVED PROMPT, KEY IMPRO
Chess Verdict
Claude skill that gives a verdict on a chess position: reads a board diagram or FEN, runs Stockfish, and retur
Run Campaign
Run a cold-outreach campaign end-to-end. Reads input JSON, processes each lead through research → profile → dr