eval-loop SKILL.md skillA

eval-loop from malloydata/publisher is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add malloydata/publisher/eval-loop@git:20261012.9b94813

Pins git:20261012.9b94813 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
malloydata/publisher
Path
skills/eval-loop/SKILL.md
Versions recorded
5
Latest version
git:20261012.9b94813, committed Oct 12, 2026
First version recorded
Sep 21, 2026
Size
31,435 bytes
Repository stars
133
License
MIT

Conduct a local Publisher evaluation loop in five steps: scrape/run, eval, diagnose, improve, checkpoint. You are the conductor: import cases into the file ledger, spawn a blind answerer, then run eval-answer, eval-diagnose, and eval-improve. Persistence is plain files: the set in the model package''s evals/ directory, runs in the set''s workdir; checkpoints are git commits of the model repo. `eval.py` runs each step from the set''s eval.toml. Use to score a model, diagnose failures, improve…

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

# The Evaluation Loop

You conduct this loop. There is no batch orchestrator to start, no eval API,
and no eval MCP tools. The ledger is plain files: the set in the model
package's git repository, and each run in the set's workdir
(`reference/ledger-schema.md` in `skill:eval-answer` defines every file and
event). `scripts/eval.py` runs each step from the set's `eval.toml`;
`reference/running-a-run.md` has the commands. Scoring is an LLM judge you spawn per case. There is no
scripted scorer, and there will not be one: a script that can pass a wrong
answer is worse than none. The scripts under `scripts/` run the loop -- they
answer, re-execute, spawn the judge, compare runs, and write the ledger -- but
none of them decides whether an answer was right.

```
scrape/run  ->  eval  ->  diagnose  ->  improve  ->  checkpoint
```

**This skill conducts; it does not restate.** Scoring lives in
`skill:eval-answer`. Components and owners live in `skill:eval-diagnose`.
Edit rules live in `skill:eval-improve`.

Do not merge **eval** into **diagnose**. A conductor who scores while
explaining writes the explanation into the score. Do not skip the **acceptance
…

Read the whole file at its exact version.

How to install

No tool to install first
claude plugin marketplace add https://markdownregistry.com/mp/a/art_z3u4f2ux3jcbwkig/4e013167a1e1ff211042964ca9c0eb714a3890b0c747dcfc0f564d8e130ebacf.json

Then /plugin install eval-loop@markdownregistry-z3u4f2ux3jcbwkig in Claude Code. It installs git:20261012.9b94813, the one graded A here, pinned to the commit it was graded at, and leaves it there.

The exact contents, by fingerprint
mdr add malloydata/publisher/eval-loop@sha256:4e013167a1e1ff21

The line at the top of this page pins the label git:20261012.9b94813, 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_z3u4f2ux3jcbwkig.svg)](https://markdownregistry.com/a/art_z3u4f2ux3jcbwkig)

1 badge image fetches in 30 days. What the badge shows.

Ask from your agent
claude mcp add --transport http markdownregistry https://markdownregistry.com/mcp/registry

Then ask it: what changed in malloydata/publisher/eval-loop since the version I have? It reads every saved version here, no key and no account. Any agent that speaks MCP can add it.

Got a copy of eval-loop from somewhere else? Ask it: where did my eval-loop come from, and what was changed? It names the public original, the version yours is closest to and every changed line. The text is not kept.

Versions

The registry has recorded 5 versions of this file, between Sep 21, 2026 and Oct 12, 2026, about one every 5 days.

Watch this file (sign in, one email per new version)

versioncommittedcommitsizeaudit
git:20261012.9b94813 latest2026-10-12 9b94813 31,435 BA view · diff
git:20261005.5a1943a2026-10-05 5a1943a 31,361 BA view · diff
git:20261001.0285b202026-10-01 0285b20 31,234 BA view · diff
git:20260923.ede90f52026-09-23 ede90f5 29,961 BA view · diff
git:20260921.b6d8d5b2026-09-21 b6d8d5b 28,893 BA view
Keep it current by itself, install once

Committed this skill to a repository? Install the markdownregistry skill updates app on it, or on all of them. When the registry saves a newer version, a pull request brings your copy to it within minutes, the newest version first, then what changed and any check it newly trips. A copy you edited is never touched, and nothing is deleted.

Install on GitHub

It asks to read and write contents and pull requests, and to read metadata. No workflow file, no Actions setting, nothing runs on your runners. Uninstall it and everything the registry kept about your repository is deleted.

Keep it current by itself

Committed this skill to a repository? Add one workflow file, and when a newer version is saved, a pull request brings your copy to it, the newest version first, then what changed and any check it newly trips. A copy you edited is never touched, and nothing is deleted.

mkdir -p .github/workflows && curl -fsSL https://markdownregistry.com/ci/skill-updates.yml -o .github/workflows/skill-updates.yml

It runs daily on GitHub Actions, no key and no account. Read the workflow before you add it.

Check a whole project

From the root of your project, one line hashes each installed skill's SKILL.md, your AGENTS.md and your CLAUDE.md, and asks the registry once. Only the fingerprints and paths are sent.

sha256sum .claude/skills/*/SKILL.md AGENTS.md CLAUDE.md | curl -s --data-binary @- https://markdownregistry.com/api/v1/check-copies

In CI, add --fail-with-body: the answer is 409 when any file is behind, so the step fails and still prints which files.

sha256sum .claude/skills/*/SKILL.md AGENTS.md CLAUDE.md | curl -s --fail-with-body --data-binary @- https://markdownregistry.com/api/v1/check-copies

No sha256sum on your machine? shasum -a 256 prints the same lines.

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (31435 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

GitHub

malloydata/publisher · 133 stars · license MIT · pushed 2026-11-17 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_z3u4f2ux3jcbwkig
GET https://markdownregistry.com/api/v1/resolve?ref=malloydata/publisher/eval-loop
GET https://markdownregistry.com/api/v1/blob/4e013167a1e1ff211042964ca9c0eb714a3890b0c747dcfc0f564d8e130ebacf

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.

More from malloydata/publisher

fix-scan-finding SKILL.md skill
malloydata/publisher · .claude/skills/fix-scan-finding/SKILL.md · Fix a CRITICAL Trivy finding that is failing CI in this repo (a vulnerability, misconfiguration, or secret from…
git:20261001.a42ad3a · audit A · 133 stars
publisher-release SKILL.md skill
malloydata/publisher · .claude/skills/publisher-release/SKILL.md · Cut a Malloy Publisher release, and write the RELEASE_NOTES.md entries a release ships. Use when asked to release, cut…
git:20261116.3b6d4a1 · audit A · 133 stars
AGENTS.md AGENTS.md file
malloydata/publisher · AGENTS.md
git:20261117.320a806 · audit B · 133 stars
CLAUDE.md CLAUDE.md file
malloydata/publisher · CLAUDE.md
git:20261116.660817a · audit B · 133 stars
AGENTS.md@examples/html-data-app AGENTS.md file
malloydata/publisher · examples/html-data-app/AGENTS.md
git:20261105.0bd58a0 · audit B · 133 stars
analyst-conventions SKILL.md skill
malloydata/publisher · examples/storefront/agents/analyst/skills/analyst-conventions/SKILL.md · House conventions for revenue questions on the storefront package. Read before answering any revenue, sales, or…
git:20261012.9b94813 · audit A · 133 stars
malloy-phrase-detection SKILL.md skill
malloydata/publisher · examples/storefront/skills/malloy-phrase-detection/SKILL.md · How to turn a storefront question into get_context search targets, including the shop vocabulary that does not appear…
git:20261012.9b94813 · audit A · 133 stars
storefront-conventions SKILL.md skill
malloydata/publisher · examples/storefront/skills/storefront-conventions/SKILL.md · What storefront's measures actually count, and the four places a reasonable reading of them is wrong. Read before…
git:20261012.9b94813 · audit A · 133 stars
AGENTS.md@packages/create-malloy-package/templates AGENTS.md file
malloydata/publisher · packages/create-malloy-package/templates/AGENTS.md
git:20261111.2af71bc · audit A · 133 stars
conventions SKILL.md skill
malloydata/publisher · packages/server/tests/fixtures/skills-agents-test/agents/analyst/skills/conventions/SKILL.md · House rules for the analyst.
git:20261012.9b94813 · audit B · 133 stars
house-rules SKILL.md skill
malloydata/publisher · packages/server/tests/fixtures/skills-agents-test/skills/house-rules/SKILL.md · How this package defines its terms.
git:20261012.9b94813 · audit B · 133 stars
malloy-analysis SKILL.md skill
malloydata/publisher · packages/server/tests/fixtures/skills-agents-test/skills/malloy-analysis/SKILL.md · The package's own analysis guide.
git:20261012.9b94813 · audit B · 133 stars

Every file in malloydata/publisher

Other files named eval-loop

eval-loop SKILL.md skill
lokeshrana9999/shipyard · plugins/shipyard-delivery/skills/eval-loop/SKILL.md · Improves prompt, skill, or instruction text in scored rounds against locked eval cases, with a frozen grader hash…
git:20261104.05cf5dd · audit A · 4 stars

Browse by kind, by grade A, or by owner.