judge-verification

name	judge-verification
description	Independent LLM judge evaluates task completion separately from the executing agent, catching false success claims by reviewing task goal, actions taken, final state, and evidence. Produces PASS/FAIL with confidence score and reasoning.
version	1.0.0
model	sonnet
invoked_by	both
user_invocable	false
tools	["Read","Write","Bash","TaskUpdate","TaskGet"]
agents	["qa","developer","master-orchestrator","planner","code-reviewer"]
category	Validation & Quality
tags	["verification","judge","llm-evaluation","completion-gate","quality-assurance"]
error_handling	strict
source	builtin
trust_score	100
provenance_sha	384542540f4ecdf3

Judge Verification

Overview

An independent LLM evaluation layer that verifies whether a task was genuinely completed. The judge reviews the original task goal, the sequence of actions taken, the final state of relevant artifacts, and the claimed completion evidence — then produces a PASS/FAIL verdict with a confidence score and actionable reasoning.

This skill is distinct from verification-before-completion: that skill runs checklist gates within the same agent context. Judge-verification uses a fresh, independent perspective with no access to the executing agent's prior reasoning, catching hallucinated success claims.

When to Use

Skill({ skill: 'judge-verification' });

Invoke when:

• Task claims completion but no test output, diff, or artifact is visible
• High-stakes tasks (security fixes, data migrations, API changes) require independent sign-off
• An agent's completion metadata seems inconsistent with the work described
• The verification-before-completion skill passes but human review is not available
• Loop detection force-done fired (partial completion needs judging)

The Iron Law

JUDGE IS INDEPENDENT — NO SHARED CONTEXT WITH EXECUTING AGENT

The judge must receive only: (1) the original task goal, (2) the list of actions, (3) the final file states. Never pass the executing agent's reasoning or internal notes to the judge.

Judge Evaluation Framework

The judge reviews four dimensions and produces a combined verdict:

Dimension 1: Task Goal Alignment

Question: Does the final state match what the task requested?

Evidence to check:

• Original task subject/description from TaskGet({ taskId })
• Files or outputs explicitly mentioned in the task
• Expected behavior described in the task goal

Score: 0-25 points

Commands:

# Get the original task goal
# (replace task-ID with actual task ID)
node -e "const fs=require('fs');const tasks=JSON.parse(fs.readFileSync('.claude/context/runtime/tasks.json','utf8')||'[]');const t=tasks.find(x=>x.id==='{{TASK_ID}}');console.log(JSON.stringify(t?.subject||'not found'));"