judge

judge is an agent for Claude Code from elixpo/claudeOps.elixpo. It costs 38 tokens per session (447 once invoked), scanned A, a copy of ensemble-judge, MIT.

A reviewer that compares two to five competing implementations and chooses the best one. It first checks whether each solution builds, passes tests, meets requirements, and avoids security problems, then scores the survivors.

In plain words
What is it for?
Use it after parallel development or when several implementations solve the same problem. It evaluates correctness, simplicity, testability, maintainability, performance, and defensive handling.
Why use it?
It provides a structured way to choose between different coding approaches instead of relying on an informal preference.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter.

Good fit Use it after parallel development or when several implementations solve the same problem. It evaluates correctness, simplicity, testability, maintainability, performance, and defensive handling.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/elixpo/claudeops.elixpo/judge
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/elixpo/claudeOps.elixpo

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.

agentmods badge for judge

README.md
[![agentmods](https://agentmods.dev/badge/agents/elixpo/claudeops.elixpo/judge/github.svg)](https://agentmods.dev/agents/elixpo/claudeops.elixpo/judge)
Your own site
<a href="https://agentmods.dev/agents/elixpo/claudeops.elixpo/judge"><img src="https://agentmods.dev/badge/agents/elixpo/claudeops.elixpo/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.

agentmods 80×15 button for judge

Your own site · 80×15
<a href="https://agentmods.dev/agents/elixpo/claudeops.elixpo/judge"><img src="https://agentmods.dev/badge/agents/elixpo/claudeops.elixpo/judge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 447 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin 95% copy Near-identical to another mod 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00038 $0.00447
Opus 5 $0.00019 $0.00224
Sonnet 5 $0.00008 $0.00089
Haiku 4.5 $0.00004 $0.00045

Measured 10d ago against content hash 41757d42f730, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

judge scanned grade A with 0 findings 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 10d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

95% identical to ensemble-judge — 2 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

agents/judge.md · 48 lines

What it actually says

You are an impartial judge evaluating competing implementations. You do NOT implement — you only evaluate and select.

Input

You will receive 2-5 competing solutions (code, diffs, or branch names).

Evaluation Process

Step 1: Binary Gates (pass/fail — any failure = elimination)

  • Compiles/builds without errors
  • All existing tests pass
  • No security vulnerabilities (injection, auth bypass, leaked secrets)
  • Meets the stated requirements

Step 2: Scoring (1-10 each, for solutions that passed gates)

Criterion Weight Description
Correctness 3x Handles all edge cases, no subtle bugs
Simplicity 2x Least complexity that solves the problem
Testability 2x Easy to test, mockable boundaries
Maintainability 1x Future developer can understand and modify
Performance 1x No unnecessary allocations, O(n) vs O(n^2)
Defensiveness 1x Error handling, input validation, fail-safe

Step 3: Verdict

WINNER: Solution [N]
SCORE: [weighted total]
RUNNER-UP: Solution [M] (score: [X])
KEY DIFFERENTIATOR: [the one thing that made the winner better]
CHERRY-PICK: [any specific element from a losing solution worth incorporating]

Rules

  • Never declare a tie — force a decision with stated reasoning
  • If the best solution has a weakness from a losing solution's strength, recommend cherry-picking
  • Run tests on each solution if possible — measured correctness beats judged correctness
  • Scoring must be independent per solution — evaluate each in isolation before comparing
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.

  1. 10d ago First seen · 48 lines · 38 tokens per session scan A 41757d42f730

Subscribe to this mod's changes

judge is an agent published in the GitHub repository elixpo/claudeOps.elixpo (2 stars, last pushed 1mo ago), licensed MIT. It adds 38 tokens to every session and 447 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to ensemble-judge, differing in 2 lines, and is treated as a copy.

Related

Other agents, from other repositories

architecture-analyst

Analyzes system architecture, identifies patterns/anti-patterns, and provides strategic recommendations. Use for architectural reviews, refactoring planning, or system design decisions.

travisjneuman/.claude · 36 tokens

deep-code-reviewer

Thorough 6-aspect code review covering correctness, security, performance, maintainability, testing, and documentation. Use for comprehensive PR reviews or code quality audits.

travisjneuman/.claude · 38 tokens

security-auditor

Use when reviewing security-sensitive code paths or running OWASP / supply-chain checks. Dispatched by code-review-loop on sensitive paths (auth, payments, crypto, users, sessions, tokens). Returns findings with severity (Critical / High / Medium / Low) and OWASP category. Context: A diff touches the auth middleware.…

duthaho/claudekit · 164 tokens

code-reviewer

Use when reviewing a diff or PR for structural issues, error handling, edge cases, complexity, and style. Dispatched primarily by code-review-loop. Returns structural findings with file:line citations and ranked severity. Pairs with security-auditor for sensitive paths. Context: A PR is ready for first-pass review.…

duthaho/claudekit · 159 tokens

spec-fidelity-auditor

Conformance of a change to its own specification (ticket, acceptance criteria, spec KB, solution plan, resolved decisions): a per-change requirements-traceability matrix swept both ways. Forward: every criterion, exact string, event name, numeric value, and named exclusion delivered, with evidence at the promised…

waqas1412/claude-harness · 186 tokens

data-flow-timing-auditor

Cross-file data-flow timing and staleness: signals read before they settle, proxy gates (timers or render flags standing in for data-readiness), one-shot consumers (analytics, seeds, redirects, caches, queue acks) snapshotting eventually-consistent state, init/hydration order, missed-event races, non-convergent…

waqas1412/claude-harness · 145 tokens