Finding Judge — Development agent for Claude Code
nightshift's Tier 2 for the reviewer.
How to install Finding Judge
Installs to ~/.claude/agents/chrispycode21-nightshift-finding-judge.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/ChrispyCode21/nightshift/HEAD/agents/finding-judge.md -o ~/.claude/agents/chrispycode21-nightshift-finding-judge.md Restart Claude Code, or start a new session, for it to be picked up.
What Finding Judge does
name: finding-judge description: nightshift's Tier 2 for the reviewer. Decides what happens to each review finding (fix, drop, hold, out of scope, standards) from a bounded packet, and adds what the reviewers missed. Started by the review-prs skill with the paths of review.md, diff.patch and issue.md. model: opus tools: Read, Write, Glob, Grep
You decide what to do about review findings; you don't edit code and you don't run commands.
You're given a run folder with:
review.md: the
Alternatives in Development
- Livemsg Gate — セッション間メッセージの主張を裏取りして SEND / HOLD を返す read-only gate 3.1k ★
- Finding Checker — Blind adversarial checker — given ONLY a finding artifact and its evidence (never the author's reasoning), tri 348 ★
- Qrspi Code Simplifier — Identifies opportunities to simplify code while preserving all functionality 38 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Green TDD
Green TDD hardener. After the builder makes red tests green, adds edge case tests and regression guards. Cheap
Ingest Concept Linker
Pass H cross-link pass of the 9-pass ingestion protocol. Reads the mechanical concept-candidates payload and d
Vouch Judge
Isolated grounding judge for Vouch. Given a verification packet file (judge instructions, corpus evidence, num
Edge Case Adversary
Independently brainstorm the edge cases a builder likely MISSED for a given deliverable, countering the "bound
Refute Verifier
Clean-context skeptic that tries to REFUTE each draft finding or claim it is given — verdicts CONFIRMED / REFU
Code Cleaner
Python code cleaning specialist. Removes excessive comments, applies DRY principles, adds professional docstri
Related Skills
Trackiq Amazon Rank Readiness
TrackIQ: Amazon Rank Readiness — an Agent Skill that decides which search terms deserve PPC or seeding spend o
AI Debate Club
A multi-agent debate engine. Two different LLMs argue any motion, research the web for themselves, and cite th
Review All Day
Time-window diff review sweep — scan every commit in a configurable merge-window-aligned span, classify each d