Evaluate Whetstone skills from session evidence when asked to compare procedure-following results or optimization candidates.
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
# eval-skills
Read the [Codex adaptation](../claude-command-compatibility.md), then read the complete [maintained workflow](../../../.claude/commands/eval-skills.md) before acting. Apply that workflow to the current request with its scope and approval boundaries intact.
Use the text following `$eval-skills` as the workflow's `$ARGUMENTS`. When selected from a natural-language request, derive arguments only from the user's stated scope; apply the workflow's missing-argument behavior for anything unspecified.
The line at the top of this page pins the label git:20260907.228ec83, 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.
GET https://markdownregistry.com/api/v1/artifacts/art_3w5lwnq73n6kkiuk
GET https://markdownregistry.com/api/v1/resolve?ref=iliaal/whetstone/eval-skills
GET https://markdownregistry.com/api/v1/blob/ea17da047ad5d3b3c38995699caf421b326bd46a3d8cf4a890b65e93e121c9a6
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.
iliaal/whetstone · .agents/skills/audit-plugin/SKILL.md · Audit Whetstone skills, agents, and commands when asked for a plugin quality or consistency audit.
iliaal/whetstone · .agents/skills/diagnose-negatives/SKILL.md · Investigate negative session feedback for a Whetstone skill when asked to diagnose its failures.
iliaal/whetstone · .agents/skills/skillopt/SKILL.md · Prepare or run Whetstone SkillOpt optimization when asked to optimize a process skill using SkillOpt.
iliaal/whetstone · .agents/skills/sync-from-repos/SKILL.md · Compare Whetstone with reference repositories when asked to sync cross-repository improvements.
iliaal/whetstone · .agents/skills/write-skill/SKILL.md · Author a new Whetstone plugin skill when asked to write a skill without upstream skills.sh sources.
iliaal/whetstone · .claude/skills/skill-distiller/SKILL.md · Fetches top-rated skills from skills.sh, analyzes them, and synthesizes one token-efficient skill combining the best…
dzhng/skills · skills/authoring/eval-skills/SKILL.md · Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each…
dzhng/duet-agent · .agents/skills/eval-skills/SKILL.md · Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each…
transitive-bullshit/skills · skills/eval-skills/SKILL.md · Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each…
transitive-bullshit/skills · provenance/upstream/eval-skills/skill/SKILL.md · Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each…