evolve-check
Skill16 starsChanged 7 months agoMITView on GitHub
Reads skill_scores.md and surfaces which skills are ready for /evolve. Analysis only — no patching, no file changes. Triggers on \"evolve check\", \"check evolve\", \"which skills need evolving\", \"evolve status\", \"skill health\".
1 other project uses a version of this.
Adopted by
Most recently taken up by
- pillumina/ascend-sleuththeir own version, first seen Nov 2, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- pillumina/ascend-sleuth★ 5same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository45 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/agents.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- actSkill
- security-checkSkill
- environment-checkSkill
- run-verificationSkill
- refactorSkill
- debug-session-archivedSkill
- generate-guardsSkill
- check-anthropicSkill
- code-reviewerSkill
- code-reviewSkill
- debug-sessionSkill
- end-sessionSkill
- evolveSkill
- fix-bugSkill
- forgetSkill
- frontend-designSkill
- guardSkill
- kit-healthSkill
- learnSkill
- map-codebaseSkill
- modeSkill
- parallel-prototypesSkill
- planSkill
- powershell-safeSkill
- product-riskSkill
- progress-reportSkill
- prototype-hypothesisSkill
- recallSkill
- refactoring-uiSkill
- scan-codebaseSkill
- scan-conventionsSkill
- search-firstSkill
- search-memorySkill
- shadow-codeSkill
- skill-creatorSkill
- smart-resumeSkill
- start-sessionSkill
- strategic-compactSkill
- tourSkill
- ux-heuristicsSkill
- verification-loopSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Allowed Tools: Read, Glob
- Very specific: line 30
Important: Only unpatched Y entries (Skill Patched = -) count toward URGENT/WATCH thresholds.
- Distinctive: line 69
Read the Last Used column.
- Distinctive: line 75
If user picks b: delete the skill file and remove its row from skill_usage.md.
- Distinctive: line 76
Do NOT silently leave stale skills — unused skills add noise to the system prompt and confuse trigger matching.
- Distinctive: line 78
Stop: Do not patch anything else.
- Distinctive: line 85
Run at End Session to catch urgent skills before the session closes
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Skill: evolve-check
Trigger: "evolve check", "check evolve", "which skills need evolving", "evolve status", "skill health"
Description: Reads skill_scores.md and surfaces which skills are ready for /evolve. Analysis only — no patching, no file changes.
Allowed Tools: Read, Glob
Steps
- Read
tasks/skill_scores.md— load all data rows. If the table is empty, report: "No skill scores logged yet. Run /learn at the end of sessions to populate skill_scores.md." Then stop. Group by skill name. For each skill compute:
- Total fires (all rows for this skill)
- Unpatched Y count — Y rows where
Skill Patched = -(not yet fixed by /evolve) - Patched Y count — Y rows where
Skill Patched != -(already fixed by /evolve) - N count (Correction Needed = N)
- Consecutive unpatched Y streak — count from the most recent row backwards until an N or a patched-Y is hit
- Whether any unpatched Y entries have "INSUFFICIENT DATA" or vague "What Failed" (no step number)
Important: Only unpatched Y entries (
Skill Patched = -) count toward URGENT/WATCH thresholds. Patched entries are resolved — they don't count as open failures.- Classify each skill:
- 🔴 URGENT — 3+ unpatched Y scores, OR 2+ consecutive unpatched Y scores on the same step
- 🟡 WATCH — exactly 2 unpatched Y scores (threshold met, ready for /evolve)
- 🔵 PATCHED — has Y entries but ALL have
Skill Patched != -; skill was fixed but not yet confirmed stable - 🟢 STABLE — 0 or 1 unpatched Y score, or 10+ consecutive N scores (includes post-patch N scores)
- ⚠️ DATA MISSING — has unpatched Y entries but "What Failed" is vague or marked INSUFFICIENT DATA
Output the report:
=== Evolve Check === 🔴 URGENT — patch now [skill] — [unpatched Y count] open failures, [streak] consecutive on Step [N] Latest failure: "[verbatim What Failed entry]" 🟡 WATCH — ready for /evolve [skill] — 2 open Y entries ([dates]) Failures: Step [N] / Step [N] 🔵 PATCHED — awaiting confirmation [skill] — all Y entries patched on [date], [N] uses since patch Next N score → promote to 🟢 STABLE 🟢 STABLE [skill] — [N count] clean sessions ⚠️ DATA MISSING — cannot evolve yet [skill] — Y logged but "What Failed" too vague to patch Action: next failure, log as: Step [N]: produced [X], needed [Y]. Fixed by [Z]. --- [N] skills flagged (🔴/🟡 only — 🔵 PATCHED are resolved pending confirmation). Run /evolve on flagged skills now? [y/N]Check
tasks/skill_usage.mdfor stale skills: Read the Last Used column. For any skill where Last Used date is more than 10 sessions ago OR shows "unknown":"[skill] hasn't been used in 10+ sessions. Options: a) Keep — schedule it for next relevant session b) Deprecate — delete SKILL.md and remove from skill_usage.md c) Skip for now"
If user picks b: delete the skill file and remove its row from skill_usage.md. Do NOT silently leave stale skills — unused skills add noise to the system prompt and confuse trigger matching.
- Stop. Do not patch anything else. Do not modify any files beyond what the user explicitly confirms above.
Notes
- This is a read-only diagnostic — it surfaces signals, never acts on them
- Run at End Session to catch urgent skills before the session closes
- /evolve does the actual patching — this just tells you when to run it
- A skill with alternating Y/N is not stable — the previous fix didn't hold; flag it manually
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

