agentleFS
Sign inStart free
BadTechBanditBadTechBandit/skills/hermes-skill-hygiene-review/SKILL.md

hermes-skill-hygiene-review

Skill4 starsChanged 7 months agoMITView on GitHub

Review active skills a Hermes agent created over time, then surface only questionable archive or merge candidates so the live skills directory stays useful instead of filling with stale one-off fixes. Use when: "review our skills", "audit stale skills", "what should we archive", "skill hygiene", "clean up agent-created skills", or any request to identify active skills that probably do not deserve to stay loaded.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 11 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Keep repeatable workflows as active skills.

    line 12
  2. Distinctive:

    If you are not confident a skill belongs to the agent-created subset, skip it.

    line 22
  3. Distinctive:

    Read the skill file.

    line 26
  4. Distinctive:

    Check recent session history only if recurrence is unclear.

    line 27
  5. Distinctive:

    Check memory or Honcho only if ongoing relevance is still unclear.

    line 28
  6. Distinctive:

    Do not pull in extra sources unless the decision is stuck.

    line 29
  7. Distinctive:

    Do not list obvious keepers.

    line 42
  8. Distinctive:

    Do not recommend deletion.

    line 51
  9. Distinctive:

    Do not move, patch, archive, or delete anything.

    line 52
  10. Distinctive:

    Leave the human in the loop.

    line 53
  11. Distinctive:

    If nothing is questionable:

    line 57
  12. Distinctive:

    If something is questionable, reply in this format:

    line 62
  13. Very specific:

    Keep the report short. Only list skills that are actually in question.

    line 72
  14. Very specific:

    Only list skills that are actually in question.

    line 73

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Hermes skill hygiene review

Review only the active custom skills a Hermes agent created over time. Surface only questionable archive or merge candidates. Touch nothing automatically.

First principles

  • Keep repeatable workflows as active skills.
  • Store reusable facts or lessons in memory.
  • Archive one-off incidents instead of leaving them active.

Review scope

  • Inspect active skills under ~/.hermes/skills/.
  • Ignore official, community, or package-installed skills.
  • Ignore explicit user-curated long-term workflow skills.
  • Ignore anything already under ~/.hermes/skills-archive/.
  • If you are not confident a skill belongs to the agent-created subset, skip it.

Evidence order

  1. Read the skill file.
  2. Check recent session history only if recurrence is unclear.
  3. Check memory or Honcho only if ongoing relevance is still unclear.
  4. Do not pull in extra sources unless the decision is stuck.

Review questions

Ask only:

  1. Did this come up more than once?
  2. Is it tied to an ongoing project or system?
  3. Is it a reusable procedure instead of a one-off diagnosis?

Surface rule

Surface a skill only if two or more answers are false or unclear. Do not list obvious keepers.

Recommendation rule

Use only:

  • archive candidate
  • merge candidate

Do not recommend deletion. Do not move, patch, archive, or delete anything. Leave the human in the loop.

Output

If nothing is questionable:

  • for manual use, reply exactly: No questionable Hermes-created skills this week.
  • for scheduled cron use, reply exactly: [SILENT]

If something is questionable, reply in this format:

Skills in question this week

1. /full/path/to/skill
   Recommendation: archive candidate|merge candidate
   Why: one short reason

Keep the report short. Only list skills that are actually in question.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free