Agent skill

retro-create

Post-invocation retrospective. Tracks skill issues and applies fixes automatically.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/retro-create

SKILL.md

Purpose

After any skill completes, review the session for issues and improve the skill to prevent recurrence.

Trigger

Run automatically after every skill invocation completes (success or failure).

What to look for

  1. Failed commands — commands that errored and required retrying or workarounds
  2. User corrections — the user stepped in to redirect, correct, or override
  3. Wrong assumptions — the skill assumed something that was not true (versions, paths, flags)
  4. Missing steps — manual steps the user had to perform that the skill should have handled
  5. Outdated instructions — references to APIs, flags, or behaviors that have changed

Protocol

For each issue found:

  1. Log — append an entry to <skill-dir>/retro.md (create the file if it does not exist)
  2. Root-cause — identify which file (SKILL.md, reference, template) contains the gap
  3. Fix immediately — edit the skill file directly to prevent recurrence. This is the default action.
  4. Log only if unsafe — record the issue with status: open only when the fix would change skill behavior in ways that cannot be verified in the current session. Require a resolution-hint on every open entry.

After logging any status: open entry, append:

Open issues remain. Run /retro-resolve to process them.

retro.md entry format

## YYYY-MM-DD — <brief title>

- **type**: failed-command | user-correction | wrong-assumption | missing-step | outdated-instruction
- **skill**: <skill-name>
- **context**: <what happened>
- **root-cause**: <why it happened>
- **fix**: <what was changed> | `status: open` if not yet fixed
- **resolution-hint**: <what the fix should look like> (required for open entries)
- **files-changed**: <list of files edited, or "none">

Constraints

  • Do not modify skill behavior or semantics — only fix gaps, errors, and omissions
  • Keep entries concise (3-5 lines per field max)
  • Do not duplicate fixes already present in skill files
  • If a fix would change skill behavior significantly, log as status: open and note the concern
  • Prefer fixing over logging. If you can describe the fix in resolution-hint, you can probably just apply it.
  • Never log as status: open simply because the session is long or context is low.

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results