Add break-ai-fix-loops skill 🤖🤖🤖

[twoicewoo]

Pull Request Checklist

• I have read and followed the CONTRIBUTING.md guidelines.
• I have read and followed the Guidance for submissions involving paid services.
• My contribution adds a new instruction, prompt, agent, skill, workflow, or canvas extension file in the correct directory.
• The file follows the required naming convention.
• The content is clearly structured and follows the example format.
• I have tested my instructions, prompt, agent, skill, workflow, or canvas extension with GitHub Copilot.
• I have run npm start and verified that README.md is up to date.
• I am targeting the main branch for this pull request.

---

Description

Adds break-ai-fix-loops, a tool- and language-neutral skill for stopping repeated AI debugging attempts and replacing them with a bounded, evidence-producing repair protocol.

The skill combines:

• stable symptom fingerprints;
• a three-attempt budget that does not reset across sessions;
• a required root-cause strategy shift after repeated evidence;
• direct verification on the real execution path;
• a known-bad negative control that must exit non-zero for the intended reason; and
• rollback verification on a separate copy while the primary modified tree remains intact.

It also includes an evidence-ledger template and a standard-library Python helper for canonical symptom fingerprints.

---

Type of Contribution

• New instruction file.
• New prompt file.
• New agent file.
• New plugin.
• New skill file.
• New agentic workflow.
• New canvas extension.
• Update to existing instruction, prompt, agent, plugin, skill, workflow, or canvas extension.
• Other (please specify):

---

Validation

• PYTHONDONTWRITEBYTECODE=1 python3 -m unittest skills/break-ai-fix-loops/scripts/test_fingerprint.py -v — 5 tests passed, exit 0.
• python3 /Users/twoicewoo/.codex/skills/.system/skill-creator/scripts/quick_validate.py skills/break-ai-fix-loops — Skill is valid!, exit 0.
• npm run skill:validate — all 417 skills valid, exit 0.
• PYTHONDONTWRITEBYTECODE=1 npm start — documentation and marketplace generation completed, exit 0.
• git diff --check — exit 0.
• GitHub Copilot CLI discovered the project skill and completed an explicit read-only behavior evaluation. The response selected SHIFT_CAUSE, prohibited another edit without new evidence, and included real-path, non-zero negative-control, separate-copy rollback, and unchanged-primary-tree gates; 8/8 output assertions passed.
• A disposable known-bad copy with fingerprint whitespace normalization disabled caused the same five-test command to fail on the intended assertion with exit 1; the untouched modified tree reran 5/5 green with exit 0.
• A deterministic staged-content scan found zero secret patterns, commerce/CTA terms, binaries, archives, or Python cache files. The only new external URL is the declared metadata.source.

Additional Notes

Affiliation disclosure: I maintain twoicewoo/stop-ai-fix-loops, the repository referenced by metadata.source.

The submitted skill is MIT-licensed, technical and vendor-neutral, and independently useful without an account or external service. The contribution contains no checkout, pricing, purchase/subscription CTA, paid bundle, or third-party runtime dependency.

---

By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.

[github-actions]

🔒 PR Risk Scan Results

Scanned 4 changed file(s).

Severity	Count
🔴 High	0
🟠 Medium	0
ℹ️ Info	2

Severity	Rule	File	Line	Match
ℹ️	skill-script-touched	skills/break-ai-fix-loops/scripts/fingerprint.py	1	skills/break-ai-fix-loops/scripts/fingerprint.py
ℹ️	skill-script-touched	skills/break-ai-fix-loops/scripts/test_fingerprint.py	1	skills/break-ai-fix-loops/scripts/test_fingerprint.py

This is an automated soft-gate report. Findings indicate review targets and do not block merge by themselves.

[github-actions]

🔍 Vally Lint Results

✅ All checks passed

Scope	Checked
Skills	1
Agents	0
Total	1

Severity	Count
❌ Errors	0
⚠️ Warnings	0
ℹ️ Advisories	0

Summary

Level	Finding
ℹ️	✅ break-ai-fix-loops (2/2 checks passed)
ℹ️	✓ [spec-compliance] All 1 skill(s) are spec-compliant.
ℹ️	✓ spec-compliance: All spec checks passed.
ℹ️	✓ [valid-refs] All file references across 1 skill(s) are valid.
ℹ️	✓ valid-refs: All file references resolve to existing files within the skill directory.
ℹ️	1 skill(s) linted, 1 passed

Full linter output

### Linting skills/break-ai-fix-loops
✅ break-ai-fix-loops (2/2 checks passed)
    ✓ [spec-compliance] All 1 skill(s) are spec-compliant.
        ✓ spec-compliance: All spec checks passed.
    ✓ [valid-refs] All file references across 1 skill(s) are valid.
        ✓ valid-refs: All file references resolve to existing files within the skill directory.

1 skill(s) linted, 1 passed