CLAUDE.md@plugins/evalsA

CLAUDE.md@plugins/evals from vinta/hal-9000 is a CLAUDE.md file: the standing instructions Claude Code reads in that repo.

Install it, pinned
mdr add vinta/hal-9000/plugins/evals/CLAUDE.md@git:20261012.4205d78

Pins git:20261012.4205d78 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 14 of 14 checks passed
Kind
CLAUDE.md file
Repository
vinta/hal-9000
Path
plugins/evals/CLAUDE.md
Versions recorded
2
Latest version
git:20261012.4205d78, committed Oct 12, 2026
First version recorded
Oct 11, 2026
Size
2,003 bytes
Repository stars
141
License
MIT

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

# CLAUDE.md

Eval suites for this repo's plugins, run by `claude plugin eval`. Each case is a directory under `<plugin>/<skill>/`.

## Running

- Run suites from the repo root with this command, adding `--case <glob>` and `--runs <n>` to narrow it; a bare run does every case three times in both arms. Every run, including the old arm of an A/B run, keeps the Sonnet judge, and `--threshold 0` keeps gradient graders from failing the exit code:

  ```bash
  claude plugin eval . --eval-dir plugins/evals --judge-model sonnet --scaffold --allow-tools WebSearch WebFetch Edit --no-publish -j 8 --threshold 0
  ```
- `--case` takes one glob without braces, so give cases you run together a shared name prefix.
- To A/B a skill change, run the committed text as the old arm: `git archive HEAD skills | tar -x -C <dir>`, copy the case directories to the same path under `<dir>`, and run the same command from `<dir>` with `--trust-plugin`.
- Read each grader across the with and without arms, not the case score.

## Writing cases
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add vinta/hal-9000/plugins/evals/CLAUDE.md@sha256:d1a15e31f9cdad5f

The line at the top of this page pins the label git:20261012.4205d78, 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_3hwzwzc5mnrussjn.svg)](https://markdownregistry.com/a/art_3hwzwzc5mnrussjn)

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

Versions

The registry has recorded 2 versions of this file, between Oct 11, 2026 and Oct 12, 2026, more than one a day.

versioncommittedcommitsizeaudit
git:20261012.4205d78 latest2026-10-12 4205d78 2,003 BA view · diff
git:20261011.2ec0c432026-10-11 2ec0c43 2,003 BA view

Audit of the latest version

A  14 of 14 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Size between 200 bytes and 200 KB (2003 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

vinta/hal-9000 · 141 stars · license MIT · pushed 2026-10-20 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_3hwzwzc5mnrussjn
GET https://markdownregistry.com/api/v1/resolve?ref=vinta/hal-9000/plugins/evals/CLAUDE.md
GET https://markdownregistry.com/api/v1/blob/d1a15e31f9cdad5fc65a1c08b8d364e314fb5f55c43ece94e87ec443d282dbac

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: what a CLAUDE.md file is · keeping AGENTS.md and CLAUDE.md in sync · which agent file does what.

More from vinta/hal-9000

publish-plugins SKILL.md skill
vinta/hal-9000 · .claude/skills/publish-plugins/SKILL.md · (project) Use when editing any file under skills/ or plugins/ to bump the plugin version and check the manifests
git:20261017.cb041d1 · audit A · 141 stars
update-playbooks SKILL.md skill
vinta/hal-9000 · .claude/skills/update-playbooks/SKILL.md · (project) Use when updating install tasks in playbooks/roles/ against upstream docs — bumping a pinned version…
git:20261020.eaa1ff3 · audit A · 141 stars
AGENTS.md AGENTS.md file
vinta/hal-9000 · AGENTS.md
git:20261020.8c774da · audit A · 141 stars
CLAUDE.md CLAUDE.md file
vinta/hal-9000 · CLAUDE.md
git:20261007.23243bf · audit A · 141 stars
CLAUDE.md@dotfiles/.claude CLAUDE.md file
vinta/hal-9000 · dotfiles/.claude/CLAUDE.md
git:20261020.db296c5 · audit A · 141 stars
AGENTS.md@dotfiles/.codex AGENTS.md file
vinta/hal-9000 · dotfiles/.codex/AGENTS.md
git:20260916.b3b244c · audit A · 141 stars
audit-agent-settings SKILL.md skill
vinta/hal-9000 · skills/audit-agent-settings/SKILL.md · Use when auditing Claude Code or Codex settings and env vars against the latest docs to suggest tailored changes
git:20261020.eaa1ff3 · audit A · 141 stars
best-practices SKILL.md skill
vinta/hal-9000 · skills/best-practices/SKILL.md · Use when about to choose, configure, or refine a tool, library, config format, API pattern, or project setup, or before…
git:20261020.eaa1ff3 · audit A · 141 stars
blindspot SKILL.md skill
vinta/hal-9000 · skills/blindspot/SKILL.md · Use when the user asks for a blindspot pass or to find their unknown unknowns, or signals unfamiliarity with a domain…
git:20261020.eaa1ff3 · audit A · 141 stars
commit SKILL.md skill
vinta/hal-9000 · skills/commit/SKILL.md · Use when making any git commit. Pass the failure, false claim, or risk the changes fix as the argument; for an addition…
git:20261020.f187727 · audit A · 141 stars
difference SKILL.md skill
vinta/hal-9000 · skills/difference/SKILL.md · Compare A vs B, weigh pros and cons of A, B, C, ..., or explain the nuance between sentences — a quick answer from what…
git:20261019.fe519c2 · audit A · 141 stars
fuck-over-engineering SKILL.md skill
vinta/hal-9000 · skills/fuck-over-engineering/SKILL.md · Use when the user asks what could be deleted or reduced to simplify a codebase, says "simplify", "over-engineered", or…
git:20261020.eaa1ff3 · audit A · 141 stars

Every file in vinta/hal-9000

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