set-up-llm-evals SKILL.md skillA
This file is byte-identical to the first copy the registry indexed, one of 3 indexed copies. Same content hash, so the same checks give both the same grade.
set-up-llm-evals from velimirmueller/claude-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add velimirmueller/claude-skills/set-up-llm-evals@git:20261009.d58ee01Pins git:20261009.d58ee01 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- velimirmueller/claude-skills
- Path
skills/ai/set-up-llm-evals/SKILL.md- Versions recorded
- 2
- Latest version
- git:20261009.d58ee01, committed Oct 9, 2026
- First version recorded
- Oct 9, 2026
- Size
- 6,764 bytes
- Repository stars
- 5
- License
- MIT
Use when an LLM feature ships on vibes — golden datasets in the repo, deterministic checks first, an LLM judge with a rubric calibrated against human labels, a small CI regression gate, a cost budget and a results history that survives a model or prompt change.
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
# Set Up LLM Evals An eval is a test suite for behaviour you cannot assert with `===`. Without one, every prompt edit, model upgrade or retrieval tweak is a guess, and a regression is found by a user. This skill sets up the smallest harness that catches real regressions: a golden set in the repo, checks that run through the same seam as production, a gate in CI and a record over time. ## 1. Audit current state Read `.claude/stack-profile.md` (`languages`, `tests`, `ci`, `ai`). If absent, detect per [stack-profile.md](../../core/_shared/stack-profile.md). Change nothing yet: ```bash ls evals evals/datasets prompts config/llm.models.json 2>/dev/null grep -rln "promptfoo\|inspect_ai\|braintrust\|deepeval\|ragas" . --include=*.json --include=*.yaml --include=*.toml --include=*.ts --include=*.py 2>/dev/null | grep -v node_modules | head grep -rn "eval" .github/workflows 2>/dev/null | head ls src/platform/llm 2>/dev/null # the seam: evals call the same code path as production git log --oneline -- prompts config/llm.models.json | head # prompt and model changes that shipped without an eval ``` …
Read the whole file at its exact version.
How to install
mdr add velimirmueller/claude-skills/set-up-llm-evals@sha256:072cb091d91feebbThe line at the top of this page pins the label git:20261009.d58ee01, 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_k6nd66swqgjolikb)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 2 versions of this file, all committed on Oct 9, 2026.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261009.d58ee01 latest | 2026-10-09 | d58ee01 | 6,764 B | A | view · diff |
| git:20261009.b6d037a | 2026-10-09 | b6d037a | 6,606 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 (6764 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
velimirmueller/claude-skills · 5 stars · license MIT · pushed 2026-10-12 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_k6nd66swqgjolikb GET https://markdownregistry.com/api/v1/resolve?ref=velimirmueller/claude-skills/set-up-llm-evals GET https://markdownregistry.com/api/v1/blob/072cb091d91feebbe0f2ae2c829a7196b40f8fde42d309f14b5ac28646429d87
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 velimirmueller/claude-skills
Every file in velimirmueller/claude-skills