Eval Harness

Community

Archive228

eval-harness

Build a repeatable eval loop that grades agent output with an LLM judge, so prompt/skill changes get scored against a baseline instead of eyeballed. Reuses loopkit's verifier subagent as the grader — do not build a new one.

Overview

Publisher	Archive228
Repository	`loopkit`
Skill name	`eval-harness`
Stars	755
Forks	124
Bundled files	Instructions only
License	MIT
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
Self-contained
Everything the model needs lives in the instructions — no extra files to sync.
Open source
Published by Archive228 on GitHub. Read the source before you install it.

Installation

Install the Eval Harness AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/Archive228/loopkit/tree/main/skills/eval-harness
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/Archive228/loopkit.git /tmp/loopkit
mkdir -p .claude/skills
cp -r /tmp/loopkit/skills/eval-harness .claude/skills/eval-harness

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Eval Harness in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Eval Harness on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Eval Harness is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Eval Harness

Every prompt tweak in a long-running agent looks like an improvement in the moment. The only way to know is a graded run against fixed inputs. Loopkit already ships .claude/agents/verifier.md — that is your grader. Do not rebuild it.

The three-stage loop


inputs.jsonl  →  runner  →  outputs.jsonl  →  verifier (per row)  →  verdicts.jsonl  →  diff vs baseline

Each stage writes to disk. No stage holds the whole run in context.

Stage 1 — inputs.jsonl

One JSON object per row: {"id": "case-01", "input": "...", "expected": "..."}.

20-100 cases is enough for a signal. More is nice, not required.
Include known-hard cases, edge cases, and a couple of trivial ones as sanity anchors.
Freeze the file. Rev the eval with a suffix (inputs-v2.jsonl) when you change it. Never edit in place — you lose the baseline.

Stage 2 — runner

A dumb loop: for each row, call the model with the current prompt/skill, capture output, write {"id": ..., "output": ...} to outputs.jsonl. No grading here — just capture.

Same temperature every run (usually 0 for evals).
Same seed / model version.
Log the git SHA of the prompt/skill under test in the file header.

If the runner is smart it will bias the eval. Keep it dumb.

Stage 3 — verifier

Fan out one subagent per row (see subagent-fanout). Each gets:

The input.
The expected output (or spec).
The actual output.
The verifier system prompt from .claude/agents/verifier.md.

Verifier returns strict JSON: {"pass": bool, "why": "..."}. Collect into verdicts.jsonl.

Diff vs baseline

Two runs of the same eval on two prompt versions → compare pass rates per case. What matters:

Overall pass rate — the headline.
Regressions — cases that were green and went red. These block ship.
New passes — cases that were red and went green. These justify ship.
Flappy cases — inconsistent across reruns. Investigate; may be genuine model nondeterminism or a bad case.

A change that raises the mean but adds regressions is usually a loss — the new failures are cases you already knew worked.

Red flags

Grader is the same model that produced the output, with the same prompt. Self-grading is lenient. Use a different persona at minimum; ideally a different model tier.
Eval passes 100% on day one. The cases are too easy, or the grader is a rubber stamp. Add adversarial cases.
Eval takes >30 minutes. Fan out. A serial 100-case eval is a serial 100-case bottleneck.
Grader sees your prompt under test. It will grade what you wanted, not what happened. Feed it only spec + input + output.
Baseline lost. Without baseline, "improvement" is vibes. Commit verdicts.jsonl to git.

When NOT to do this

One-off script — build a checklist, not a harness.
Prompt that changes daily and won't stabilize — evals need a fixed target.
Task where "correct" isn't checkable (open-ended creative writing) — use human eval or a rubric-based grader, not pass/fail.

The verifier is already yours. The harness is 100 lines of glue around it.

Frequently asked questions

What does the Eval Harness AI skill do?

Why use Eval Harness on TypingMind?

Because you install it once and use it with any model. Eval Harness is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Eval Harness in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Archive228/loopkit/tree/main/skills/eval-harness. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Eval Harness?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Eval Harness?

As many as you like. As long as a model supports skills, you can use Eval Harness with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Eval Harness AI skill free?

Yes. It is published on GitHub by Archive228 under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Release Openclaw Plugin Testing

openclaw

OrganizationPopular

Plan and run pre-release OpenClaw plugin validation across bundled plugins, package artifacts, lifecycle commands, doctor/fix, config round-trip, gateway startup, SDK compatibility, Docker E2E,…

Solid

Write, port, review, or test Solid 2 UI in OpenClaw (Control UI, plugin views, Canvas hosts), including state ownership, Lit interop during the migration, native overlays, testing, performance, and…

Auto Qa

Run OpenClaw-wide autonomous QA and live/stress campaigns across independent subsystem lanes, with verified fixes and a resumable evidence report.

Autoreview

Structured code review when explicitly requested, preferring OpenAI/Codex before Claude.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Eval Harness

The three-stage loop

Stage 1 — inputs.jsonl

Stage 2 — runner

Stage 3 — verifier

Diff vs baseline

Red flags

When NOT to do this

Frequently asked questions

Related AI Skills

Set up your own AI workspace now