harness-eval

harness-eval is a skill for Claude Code from uppifyagency/claude-harness. It costs 74 tokens per session (423 once invoked), scanned A, original, MIT.

A standalone quality check for a running application against a written specification. It sends an evaluator to test the application and reports a verdict, scores, and critical issues.

In plain words
What is it for?
Use it for requirements testing, QA, final checks before release, or comparing different builds.
Why use it?
It lets you check an existing build or a manually fixed build without running a complete development loop. It also supports comparing results with earlier evaluations.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: names the AskUserQuestion tool.

Part of the claude-harness plugin — 4 skills, 3 agents, 1 hook shipped together

Good fit Use it for requirements testing, QA, final checks before release, or comparing different builds.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/uppifyagency/claude-harness/harness-eval
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.

Any agent
npx skills add uppifyagency/claude-harness --skill harness-eval
Clone the repo
git clone --depth 1 https://github.com/uppifyagency/claude-harness

Made for: Claude Code.

Or install claude-harness, the plugin that ships this one along with the rest of its 4 skills, 3 agents, 1 hook.

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-eval

README.md
[![agentmods](https://agentmods.dev/badge/skills/uppifyagency/claude-harness/harness-eval/github.svg)](https://agentmods.dev/skills/uppifyagency/claude-harness/harness-eval)
Your own site
<a href="https://agentmods.dev/skills/uppifyagency/claude-harness/harness-eval"><img src="https://agentmods.dev/badge/skills/uppifyagency/claude-harness/harness-eval/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 harness-eval

Your own site · 80×15
<a href="https://agentmods.dev/skills/uppifyagency/claude-harness/harness-eval"><img src="https://agentmods.dev/badge/skills/uppifyagency/claude-harness/harness-eval.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 74 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 423 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 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00074 $0.00423
Opus 5 $0.00037 $0.00211
Sonnet 5 $0.00015 $0.00085
Haiku 4.5 $0.00007 $0.00042

Measured 9d ago against content hash d6f9c8d512a7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

harness-eval 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 9d 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.

skills/harness-eval/SKILL.md · 56 lines

What it actually says

Harness Eval — Standalone Evaluation

Overview

Run the evaluator agent independently to assess an application against a specification. Useful for:

  • Checking existing applications against requirements
  • Re-evaluating after manual fixes
  • Final quality gate before shipping
  • Comparing quality across different builds

Execution

Step 1: Locate the Spec

Check for spec in this order:

  1. Argument path (if provided)
  2. .harness/spec.md in current project
  3. Ask the user to provide or point to a spec

Step 2: Verify Application is Running

Check if a development server is already running. If not, look for common start commands:

  • package.json scripts (dev, start)
  • Makefile targets
  • Docker compose files
  • Ask the user if unclear

Step 3: Dispatch Evaluator

Dispatch the harness-evaluator agent with:

  • The spec file
  • Instructions to test the running application
  • Any previous eval reports (for comparison)

Step 4: Present Results

Show the user:

  • Overall verdict (PASS / ITERATE / PIVOT)
  • Quality dimension scores
  • Critical issues found
  • Comparison with previous eval (if available)

Determine the eval round number from existing .harness/eval-report-*.md files and write the new report as the next number.

Additional Resources

Reference Files

  • references/custom-criteria.md — Adapting evaluation criteria for non-visual projects
Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 9d ago First seen · 56 lines · 74 tokens per session scan A d6f9c8d512a7

Subscribe to this mod's changes

harness-eval is a skill published in the GitHub repository uppifyagency/claude-harness (5 stars, last pushed 5mo ago), licensed MIT. It adds 74 tokens to every session and 423 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.