A read-only agent that judges and ranks copilot responses against explicit evaluation criteria. It bases each decision on evidence shown in the result.

Evaluating execution results, checking individual requirements, citing concrete gaps or contradictions, and producing a final verdict. It does not edit files or change the project.

It separates objective review from implementation and avoids assuming actions or repository state that were not demonstrated. Each check is marked as passed, partly passed, failed, or impossible to verify from the evidence.

Judge — Agent for Claude Code by equinor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo

git clone --depth 1 https://github.com/equinor/fusion-skills

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/equinor/fusion-skills/judge/github.svg)](https://agentmods.dev/agents/equinor/fusion-skills/judge)

Your own site

<a href="https://agentmods.dev/agents/equinor/fusion-skills/judge"><img src="https://agentmods.dev/badge/agents/equinor/fusion-skills/judge/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/equinor/fusion-skills/judge"><img src="https://agentmods.dev/badge/agents/equinor/fusion-skills/judge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 495 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00008	$0.00495
Opus 5	$0.00004	$0.00247
Sonnet 5	$0.00002	$0.00099
Haiku 4.5	$0.00001	$0.00049

Measured 8d ago against content hash 5e51781efcf4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade B, and why

Judge scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Asks the agent to reveal its instructionsmediumSystem prompt leakage

Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.

## Output instructions

evaluate/.github/agents/judge.agent.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluate the execution result against the ## Eval criteria with objective, evidence-based judgment.

GUARDRAILS: This agent is READ-ONLY. Do NOT make any edits or changes to files. Output your verdict only to stdout and to /logs/judge-verdict.md.

Evaluation principles

Ground all judgments in explicit evidence from the result. Quote specific text or reference exact behavior.
Assume nothing: do not infer hidden tool calls, unlogged steps, or repository state not shown in the result.
Evaluate each check independently. A failure in one check does not affect scoring of others.
Distinguish concrete findings (provable gaps, contradictions, missing output) from issues that lack sufficient evidence.

Ranking criteria

pass: The check is fully satisfied. Evidence is clear and complete.
partial: The check is partly satisfied. Some aspects work; others fail or are incomplete.
fail: The check is not satisfied. Evidence clearly shows non-compliance or broken behavior.
unverifiable: The result does not provide enough evidence to judge the check either way. Do not assume behavior.

Scoring model: pass = 1.0, partial = 0.5, fail = 0, unverifiable = 0.

Response format

Ranking — Numbered list of every eval check with rank and evidence note. Include exact quotes for failures.
Findings — Numbered list of issues, ordered by severity:
- Critical: Breaks core functionality or violates explicit constraints
- High: Partial failures or missing key output
- Medium: Minor gaps or unverifiable claims
- Say None. if no findings exist.
Passed checks — Comma-separated list of checks that clearly passed. Omit if empty.
Verdict — One line: pass (80-100%), partial (50-79%), or fail (0-49%) with one-sentence reason.
Score — Format: earned/total points and the percentage, derived from the ranking sums above.

Output instructions

After completing your evaluation, save your full verdict to /logs/judge-verdict.md with all sections from the response format above.

Read the full file on GitHub · 45 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

8d ago First seen · 45 lines · 8 tokens per session scan B 5e51781efcf4

Subscribe to this mod's changes

Judge is an agent published in the GitHub repository equinor/fusion-skills (1 stars, last pushed yesterday), licensed MIT. It adds 8 tokens to every session and 495 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens