harness-evaluator SKILL.md skillA
harness-evaluator from romabeckman/harness-kit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add romabeckman/harness-kit/harness-evaluator@git:20260810.b9eb234Pins git:20260810.b9eb234 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- romabeckman/harness-kit
- Path
skills/harness-evaluator/SKILL.md- Versions recorded
- 5
- Latest version
- git:20260810.b9eb234, committed Aug 10, 2026
- First version recorded
- May 23, 2026
- Size
- 6,656 bytes
- Repository stars
- 30
- License
- MIT
Harness performance evaluator. Reads all execution traces in docs/harness-history/traces/, computes composite scores per skill_chain, identifies the Pareto frontier of best harness configurations, and recommends the optimal chain for the next session. Run periodically or on demand to guide harness optimization.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Harness Evaluator — Performance Analyzer You are a **harness performance analyst**. Your responsibility is to read the accumulated execution traces in `docs/harness-history/`, compute objective scores per harness configuration, identify the Pareto frontier, and produce actionable recommendations. --- ## ROLE Aggregate session data from `docs/harness-history/traces/` and produce a ranked comparison of harness configurations (skill chains). This is the scoring phase of the Meta-Harness loop: it converts raw traces into the signal that the `meta-harness` proposer uses to diagnose and improve skills. --- ## PRECONDITIONS 1. **Verify history exists** — check that `docs/harness-history/traces/` exists and contains at least one session. - If empty or missing: inform the user and stop. Suggest running `/harness-kit:tdd-orchestrator` first to generate sessions. 2. **Read `config.md`** — load score weights from `docs/harness-history/config.md`. - If missing: create it using the template in `harness-tracer/SKILL.md`, then proceed. 3. **Count available sessions** — list all `session-*` directories under `traces/`. …
Read the whole file at its exact version.
How to install
mdr add romabeckman/harness-kit/harness-evaluator@sha256:2a59d819de7b7fe2The line at the top of this page pins the label git:20260810.b9eb234, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_v6i3ug4nriknms4y)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between May 23, 2026 and Aug 10, 2026, about one every 20 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260810.b9eb234 latest | 2026-08-10 | b9eb234 | 6,656 B | A | view · diff |
| git:20260810.48f7be1 | 2026-08-10 | 48f7be1 | 6,256 B | A | view · diff |
| git:20260526.39179e2 | 2026-05-26 | 39179e2 | 6,269 B | A | view · diff |
| git:20260523.0a17b9f | 2026-05-23 | 0a17b9f | 6,161 B | A | view · diff |
| git:20260523.82d7505 | 2026-05-23 | 82d7505 | 6,213 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6656 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
romabeckman/harness-kit · 30 stars · license MIT · pushed 2026-11-07 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_v6i3ug4nriknms4y GET https://markdownregistry.com/api/v1/resolve?ref=romabeckman/harness-kit/harness-evaluator GET https://markdownregistry.com/api/v1/blob/2a59d819de7b7fe254a42478bbc1c8a874c26d91dd6d0b2ac11114168cf4cc21
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.