Reviewer Sonnet — Development agent for Claude Code
agent-review specialist reviewer running on the sonnet model tier.
How to install Reviewer Sonnet
Installs to ~/.claude/agents/cruglobal-agent-review-reviewer-sonnet.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/CruGlobal/agent-review/HEAD/agents/reviewer-sonnet.md -o ~/.claude/agents/cruglobal-agent-review-reviewer-sonnet.md Restart Claude Code, or start a new session, for it to be picked up.
What Reviewer Sonnet does
name: reviewer-sonnet description: agent-review specialist reviewer running on the sonnet model tier. Selected by the review skill's launch table; never invoke directly. model: sonnet effort: low
You are one lane of a multi-agent code review. Your entire assignment — role, rules, diff, output contract — arrives in the task prompt. Follow it exactly; add nothing beyond it. Keep reasoning brief and focused: verify before you report, and spend your effort on the diff, not on narrating.
Alternatives in Development
- Learning Reviewer — Internal agent for reviewing learnings and promoting them to rules or skills 150 ★
- Miner Low — Delegate worker pinned to LOW reasoning effort, the cheapest tier 31 ★
- Lead Review — Review pipeline lead 27 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Reviewer Haiku
agent-review specialist reviewer running on the haiku model tier. Selected by the review skill's launch table;
Reviewer Opus
agent-review specialist reviewer running on the opus model tier. Selected by the review skill's launch table;
Explorer Max
Explorer with a long report (≤6k characters) for table/list answers that don't fit in 1.5k. Same model as expl
Fleet Review Alt
Independent second reviewer on a different model. Use for risky changes that deserve two reviews, or when the
iOS Performance Engineer
Use this agent to review native iOS code for performance problems — SwiftUI view updates and identity, slow Li
Setup Guide
Run automatically at the start of every Lumen session to validate MCP connections and seed the context object.
Related Skills
Ready Plans
Show every plan in ready/ as a scannable table — lane (sonnet/fable/sol), priority, and whether a cloud drain
Claude Glm Stack
Run Claude Desktop on Z.AI GLM (GLM Coding Plan) via third-party 3P gateway mode + tier-rewriting local proxy:
AI Dev Team
Token-efficient multi-agent dev team for Claude Code: planner, parallel builders, devops, code reviewer, teste