task-observer: bugfix-tdd updates from 2026-08-17

[divad12]

Target

skill: .agents/skills/bugfix-tdd/SKILL.md

Observations addressed

• Observation 15 (journology/log.md): Bug Bashes Need Principle And Enforcement Capture → changed: enhanced Step 5 to capture root cause + principle (not just symptom), note current and next enforcement candidates, and promote repeated/high-blast-radius bug classes to mechanical guardrails (lint rule, regression test, shared helper, schema constraint, API contract, or docs/ai rule) rather than leaving them as learning notes
• Observation 17 (journology/log.md): Review And Bug Learning Need Clear Global/Local Layers (bugfix-tdd portion) → changed: added routing guidance to Step 5 — route fixes to the right artifact category (learning store for the principle, enhancement backlog for follow-on work)

---

Generated by Claude Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 7f2dfe8cc1

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Re-run verification after creating enforcement artifacts

When a recurring or multi-caller bug promotes a shared helper, schema constraint, API contract, or regression test here, the skill now mandates additional implementation work after Step 4 has already run the only full-suite verification; Step 6 then proceeds directly to the summary. Those post-verification changes can therefore introduce failures that are never checked, so move this promotion before Step 4 or explicitly repeat verification afterward.

Useful? React with 👍 / 👎.