aiw-planninglisted

Structured planning process for producing a code-aware implementation plan before writing code, including the pre-planning codebase baseline checks (smoke tests and the global suite, test readiness, bounded-change confirmation) and the task modality classification the rest of the plan depends on. Use this skill at the start of every task before any implementation, whether the user says 'plan' or just hands the agent an issue. It owns the codebase baseline check, the modality classification step (procedure in aiw-ground-truth), the plan's structure and required content, the oracle and verification expectations the plan must state, the issue-vs-codebase priority rule, assumption classification, bounded-change confirmation, and handling human feedback at plan review. It does not own oracle definitions or the trust hierarchy (aiw-ground-truth), test mechanics (aiw-testing), evidence and sufficiency (aiw-verification), or post-failure investigation (aiw-failure-analysis).
philippe-ths/ai-coding-workflow · ★ 0 · AI & Automation · score 66

Install: claude install-skill philippe-ths/ai-coding-workflow

# Planning Read this file when producing or revising an implementation plan at the start of a task. ## Where This Skill Sits in the Workflow Planning is step 2 of the task sequence. The plan does not reproduce the rules of the other skills; it identifies what each will need to address during implementation and defers the mechanics to them. The full sequence, for orientation: 1. **aiw-github** at task start: confirm the GitHub issue, read it and its comments, create or switch to an issue-scoped branch. 2. **aiw-planning** (this skill): establish the codebase baseline, classify the modality, name the oracle, produce the plan for human review. 3. **Implementation.** As work proceeds, aiw-ground-truth governs fixtures and oracles, aiw-testing governs test mechanics, aiw-verification governs the evidence required to declare done. 4. **The done gate** before any "done" claim: aiw-verification's justification step for whether the change is correct, aiw-validation for whether the deliverable this plan named actually exists and does what was asked, and aiw-housekeeping for whether anything the task caused should now be removed or moved. 5. **aiw-github** again for commit, push, and pull request, taken without asking once the done gate has run. 6. **aiw-failure-analysis** if a "done" claim is later contradicted; it audits, may surface a plan-level flaw, and may restart the sequence from re-planning. This skill owns step 2. The overview exists so the full sequence is visible at a g