agentleFS
blixvipblixvip/skills/backbefore/SKILL.md

backbefore

Skill4 starsChanged 3 months agoMITView on GitHub

Restore the version immediately before a rejected implementation, then reapply the user's latest request as a minimal compatible change. Use when the user invokes /backbefore or says they preferred the version before the latest change.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Treat the version immediately before the rejected change as the baseline.

    line 8
  2. Distinctive:

    Verify both that the prior behavior is restored and that the new request works.

    line 16
  3. Distinctive:

    Treat the previous version as the baseline and the latest request as the delta.

    line 20
  4. Distinctive:

    Do not merely undo the new feature.

    line 21
  5. Distinctive:

    Do not redesign, rewrite, or improve unrelated parts.

    line 22
  6. Distinctive:

    Never use a destructive full reset, checkout, or revert that could erase other work.

    line 23
  7. Distinctive:

    If several older versions exist, use the state immediately before the rejected change.

    line 24
  8. Distinctive:

    Use the fastest capable model when model switching is supported.

    line 26
  9. Distinctive:

    If genuinely blocked, return only:

    line 37

≈ “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

/backbefore

Treat the version immediately before the rejected change as the baseline. Restore its look, behavior, and structure, then implement the latest request in a way that fits that baseline.

Workflow

  1. Identify the last good state from recent context, Git history or diffs, screenshots, or prior output.
  2. Determine exactly what the rejected implementation changed or damaged.
  3. Restore only the affected parts and preserve unrelated work.
  4. Reapply the latest request as a minimal, compatible change.
  5. Verify both that the prior behavior is restored and that the new request works.

Rules

  • Treat the previous version as the baseline and the latest request as the delta.
  • Do not merely undo the new feature; integrate it better.
  • Do not redesign, rewrite, or improve unrelated parts.
  • Never use a destructive full reset, checkout, or revert that could erase other work.
  • If several older versions exist, use the state immediately before the rejected change.
  • For UI work, verify the rendered result rather than source alone.
  • Use the fastest capable model when model switching is supported.
  • Work immediately and return only after restoration and the new change are verified.

Return only:

✅ Back before: [what was restored]
✅ Still added: [how the new request now fits]
Verified: [specific check and result]

If genuinely blocked, return only:

⛔ Blocked: [exact missing history, file, or external dependency]
Needed: [single requirement to continue]

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.