Robin — Development agent for Claude Code
Read-only, skeptical evidence auditor for a Superloopy criterion.
How to install Robin
Installs to ~/.claude/agents/beefiker-superloopy-robin.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/beefiker/superloopy/HEAD/agents/robin.md -o ~/.claude/agents/beefiker-superloopy-robin.md Restart Claude Code, or start a new session, for it to be picked up.
What Robin does
name: robin description: Read-only, skeptical evidence auditor for a Superloopy criterion. Use to judge whether Superloopy's deterministic re-run actually satisfies the scenario and write a verdict. model: sonnet tools: Read, Grep, Glob, Bash
You are robin, an independent, read-only Superloopy evidence auditor. Assume the implementation is NOT done.
You are given one criterion's scenario and Superloopy's deterministic re-run artifact: Superloopy already re-ran the proof in-process and
Alternatives in Development
- Gitnexus Synthesis Critic — GitNexus final review synthesis critic 45.8k ★
- Scenario Analyst 1k ★
- Verify Claim Faithfulness — A verification agent that independently checks, at the sentence level, whether each cited source actually SUPP 134 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Cook Verify
jeff verify stage for operation tasks. Independently verify every deterministic postcondition and acceptance c
Franky
Implementation worker for one Superloopy criterion or one non-overlapping slice. Use when a bounded build/edit
Acceptance Auditor
Verifies that the work on the current issue branch satisfies every acceptance criterion written in the issue.
Casper
CASPER-3, the woman. Judges a proposition on pragmatics — whether it can actually be run, maintained, and live
Skill Applied Judge
Use this agent at session-end Phase 3.6.6 (#645 L3) to judge — from the session transcript tail — whether each
Judge Router
Decides whether a stage's verdict requires cross-vendor or same-vendor judging by calling pp_harness.gate_elig
Related Skills
Gov Metrics
Governance metrics — measure whether the gates actually work. Block rate, override/waiver rate, false-block pr
Lorenzini
Claude Code skills that wait for a third-party PR reviewer and decide whether its verdict actually means pass
Claude Interaction Tests
Test whether your CLAUDE.md actually works. A Claude Code skill that restructures agent-facing docs (CLAUDE.md
