harness-evaluator

harness-evaluator is an agent for Claude Code from AuroraPixel/ai-native-harness-skill. It costs 35 tokens per session (2,250 once invoked), scanned A, a copy of harness-evaluator, Apache-2.0.

An independent code-review agent that checks whether software changes meet specified acceptance criteria. It uses repeatable checks first, then looks more deeply for logic errors, edge cases, concurrency problems, and security issues.

In plain words
What is it for?
Use it at project checkpoints to evaluate implementations, run deterministic tests, inspect deeper behavior, and report issues with evidence.
Why use it?
It gives a separate review of a code change and flags uncertain cases for human judgment instead of guessing. It checks normal, error, and boundary paths.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions Codex; built for gstack.

Part of the ai-native-harness-skill plugin — 2 skills, 5 agents shipped together

Good fit Use it at project checkpoints to evaluate implementations, run deterministic tests, inspect deeper behavior, and report issues with evidence.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/aurorapixel/ai-native-harness-skill/harness-evaluator
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/AuroraPixel/ai-native-harness-skill

Made for: Claude Code.

Or install ai-native-harness-skill, the plugin that ships this one along with the rest of its 2 skills, 5 agents.

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 harness-evaluator

README.md
[![agentmods](https://agentmods.dev/badge/agents/aurorapixel/ai-native-harness-skill/harness-evaluator.svg)](https://agentmods.dev/agents/aurorapixel/ai-native-harness-skill/harness-evaluator)
Your own site
<a href="https://agentmods.dev/agents/aurorapixel/ai-native-harness-skill/harness-evaluator"><img src="https://agentmods.dev/badge/agents/aurorapixel/ai-native-harness-skill/harness-evaluator.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,250 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 100% 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 $0.00035 $0.02250
Opus 5 $0.00017 $0.01125
Sonnet 5 $0.00007 $0.00450
Haiku 4.5 $0.00003 $0.00225

Measured 8d ago against content hash 5a946a38040f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

harness-evaluator 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 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.

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

100% identical to harness-evaluator — 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.

plugins/ai-native-harness-skill/agents/harness-evaluator.md · 115 lines

How it starts

The opening of the file, as written. The whole thing — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluator Agent

Identity

Senior architect and QA lead performing independent evaluation of code changes against checkpoint acceptance criteria.

Behavioral Mindset

Be thorough and evidence-based. Every claim must be backed by test output, screenshots, or API responses. When uncertain whether behavior matches spec, mark as REVIEW rather than guessing. Deep logic analysis — look beyond surface patterns to find edge cases, concurrency issues, and security vulnerabilities.

Principles

  1. Evidence over opinion — every verdict must reference artifacts in evidence/
  2. Tier 1 before Tier 2 — run all deterministic checks before LLM code review
  3. Spec defines WHAT, you decide HOW — acceptance criteria say what to verify; you choose the verification method
  4. Normal + error + boundary — every flow gets at least three paths tested
  5. REVIEW over guess — if uncertain whether behavior matches spec, mark REVIEW for human
  6. Scoped judgment — evaluate only this checkpoint's changes, not the entire codebase
  7. Classify REVIEW items precisely — every review_item carries severity, auto_fixable, and requires_human_judgment so the Orchestrator can auto-resolve trivial issues without human input
  8. Artifact-shape match — when a criterion names a specific artifact (file path, screenshot, report), evidence MUST be that artifact or a same-shape facsimile — never a same-property proxy in a different shape. A POST body ≠ state.json excerpt; a source-grep ≠ dist/<name>.js grep; a fixture screenshot ≠ a popup screenshot. If the named artifact genuinely cannot be produced this iter, mark REVIEW with auto_fixable: false and name the gap — do not accept a proxy. (Pairs with Generator Principle 7 "Artifact-shape evidence" on the emit side.)
  9. Coverage-measurement gate (full-verify) — when the task includes a backend/infrastructure/fullstack checkpoint OR the spec explicitly requires measured coverage (e.g. "85% per TESTING.md"), verification-report.md MUST carry numeric coverage_percent in frontmatter. If coverage tooling is absent: set verdict: FAIL, increment hard_failures, record the missing tooling with concrete fix guidance (install command, expected output). Qualitative assessment ("test density consistent with 85%+") is informational, never a substitute for measurement. Only frontend-only tasks with no coverage requirement may set coverage_percent: N/A.

Read the full file on GitHub · 115 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.

  1. 8d ago First seen · 115 lines · 35 tokens per session scan A 5a946a38040f

Subscribe to this mod's changes

harness-evaluator is an agent published in the GitHub repository AuroraPixel/ai-native-harness-skill (11 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 35 tokens to every session and 2,250 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to harness-evaluator, differing in 2 lines, and is treated as a copy.

Related

Other agents, from other repositories

pr-test-analyzer

Use this agent when you need to review a pull request for test coverage quality and completeness. This agent should be invoked after a PR is created or updated to ensure tests adequately cover new functionality and edge cases. Examples:\n\n \nContext: Daisy has just created a pull request with new…

anthropics/claude-code · 0 tokens

ai-hygiene-auditor

Audit codebases for AI-generation warning signs: vibe coding patterns, agent psychosis indicators, slop artifacts, and Tab-completion bloat. Specialized complement to bloat-auditor.

athola/claude-night-market · 48 tokens

sap-test-plan-reviewer

Adversarial review of a test-case plan produced by design-cases. READS the actual ABAP source snapshot (plus findings.md, flow.md, units.md, and the TC-.md files) to catch branches and MESSAGEs the plan missed, checks total case count against the enumerated minimum, checks every mandatory category has at least one…

marcellourbani/vscode_abap_remote_fs · 161 tokens

edge-case-explorer

Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…

testdouble/han · 135 tokens

test-reviewer

Reviews test coverage and test quality for code changes.

ai-sdlc-framework/ai-sdlc · 13 tokens

ring:qa

Senior QA Analyst for financial systems. Supports 6 testing modes — unit (default), fuzz, property, integration, chaos, goroutine-leak. Dispatched by orchestrator with mode parameter; loads mode-specific file from qa-modes/.

LerianStudio/ring · 52 tokens