skill-evals SKILL.md skillA
skill-evals from geekfamilycorp/tricorderkit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add geekfamilycorp/tricorderkit/skill-evals@v1.0Pins v1.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- geekfamilycorp/tricorderkit
- Path
cowork-plugins/tk-knowledge/skills/skill-evals/SKILL.md- Versions recorded
- 1
- Latest version
- v1.0, committed Oct 28, 2026
- Size
- 9,283 bytes
- Repository stars
- 32
- License
- MIT
Écrit, exécute et consigne les jeux d'evals des skills critiques de l'utilisateur (déclenchement, non-déclenchement, qualité, régression, token-hygiene) au format natif de skill-creator (evals.json + triggers.json), avec contrôles déterministes hors LLM (check_static.py : frontmatter, description ≤ 1024, chemins en dur, IP, nom de modèle versionné, renvois → skill fantômes, changelog) AVANT tout juge LLM. À utiliser quand l'utilisateur dit « écris/lance les evals de X », « teste ce skill avant…
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
# skill-evals — le banc d'essai des skills critiques Un skill sans test se dégrade en silence : un renvoi vers un skill archivé, un nom de modèle qui dérive, une description qui ne déclenche plus. Ce skill impose la règle de la Phase 8 : **chaque skill critique a ≥ 5 tests, les régressions se voient, et le déterministe passe avant le juge LLM.** Règles citées : R51 (recouper, ne pas affirmer de mémoire), R53 (proposer, ne pas exécuter), R59 (repackager après chaque MAJ — donc tester avant), R31 NO-DROP. ## Quand lancer | Moment | Quoi | |---|---| | Avant toute réécriture d'un skill | jeu complet → c'est la **ligne de base** ; sans elle, pas de régression détectable | | Après la réécriture, avant repackaging (R59) | jeu complet ; comparer à la ligne de base ; un cas qui passait et échoue = STOP | | Après un renommage/archivage de skill voisin | `check_static.py` seul (renvois → skill) sur tous les skills qui le citent | | Périodiquement (tâche planifiée `default`, jamais `deep`) | `check_static.py` sur tous les SKILL.md ; les cas LLM restent manuels | ## Le jeu d'evals d'un skill — `evals/<skill>/` …
Read the whole file at its exact version.
How to install
mdr add geekfamilycorp/tricorderkit/skill-evals@sha256:d1e5761d6c78faa7The line at the top of this page pins the label v1.0, 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_mbhn2joa6vv5v7ps)
0 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 28, 2026.
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 (9283 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
geekfamilycorp/tricorderkit · 32 stars · license MIT · pushed 2026-11-01 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_mbhn2joa6vv5v7ps GET https://markdownregistry.com/api/v1/resolve?ref=geekfamilycorp/tricorderkit/skill-evals GET https://markdownregistry.com/api/v1/blob/d1e5761d6c78faa7d32a71e1b951d3785c51fe9c602fa5d4c5c3b3db1d90b617
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.