Skip to content

tibursocampos/agent-dev-toolkit / core/skills/sdd-plan

sdd-plan

Verified

Create a baby-step PLAN from an existing PRD (agent PLAN .md; language per LANGUAGE.md + preferences). Feeds sdd-develop. Use when creating a plan or invoking /sdd-plan.

references/quality-checklist.md

View on GitHub
## Quality checklist (before handoff)

- [ ] User confirmed **sim** on canonical PLAN path (`PIPELINE.md` § Confirm before write)
- [ ] Canonical PRD on disk; PRD path and `NNN` match the PLAN filename
- [ ] Body follows `templates/sdd/PLAN.md` (REQ→step map; Aceite cites REQ/CA)
- [ ] `direct`: persisted approved PRD is sufficient; steps map PRD behavior/REQ/CA with PLAN-native IDs and no fabricated checklist rows. `orchestrated`: required O2 backlog/approval/checklist gates remain enforced.
- [ ] Optional direct STORY/checklist is explicitly linked or provided and does not conflict with PRD scope, acceptance, or dependencies; conflict blocks for operator resolution.
- [ ] `## Related` present; PRD↔PLAN mutual cite when both exist; omit-if-absent (`STORAGE.md` § Navigation block / REQ-009)
- [ ] Every PRD REQ and acceptance criterion appears in some step
- [ ] Implementation DoR is reviewable before the first step: every REQ→CA→step has a named planned validation, scope/OOS and dependencies are explicit, and unresolved implementation-affecting decisions block start
- [ ] Acceptance traceability records planned validation separately from execution evidence; no planned check is represented as a passing result
- [ ] Step/US DoD requires accepted outcomes, passed declared checks, and recorded evidence; clarification `READY` is not used as authorization or completion
- [ ] PLAN preserves the develop-session gate lifecycle: finish edits/re-read/validation, record `tests_run`, update status/evidence, then reset terminally; no helper, validation, or artifact edit follows reset in that scope
- [ ] Mixed test strategy is explicit; deferred checks have a final US consolidation owner and cannot approve an intermediate step
- [ ] Final consolidation maps acceptance criteria to tests/evidence, preserves unique unit and unrelated smoke coverage, and removes only proven redundancy
- [ ] Final US/PLAN DoD requires consolidation checks and STATE/EVD validation at the configured evidence level; valid prior evidence is retained or explicitly superseded with a reason and replacement
- [ ] Vague Aceite challenged and rewritten (observable checks only)
- [ ] No file-named / task-shaped step titles (file/class/script only) — `story-sizing.md` / `anti-task-shatter.md`
- [ ] When present, story `ARCH/` / `ANALYSIS/` cited by portable path (not pasted)
- [ ] `validate-prd` and `validate-plan` exit 0 on the written paths
- [ ] Step prose in the resolved artifact language (`LANGUAGE.md`; English only when the operator requests it)
- [ ] No full implementation code blocks in the PLAN
- [ ] When the established project style differs from the toolkit style bar, the PLAN records that difference and the open question (project vs toolkit). Implementation does not restyle existing members until the operator answers. An order to keep the project pattern closes the question
- [ ] No full PRD / memory-bank dump in PLAN or handoff prompts
- [ ] PLAN magro: SQL/DDL/JSON/OpenAPI omitted **only** if a canonical path already exists (bank phase 2 or `ARCH/` / `ANALYSIS/`); otherwise create that file first and cite the path
- [ ] Output path: `features/**/PLAN/PLAN_NNN_*.md` or global `.../features/**/PLAN/` only (not root `PLAN/`, ad-hoc `docs/`, or `{{TOOLKIT_ROOT}}/` outside `sdd/.../features/`)
- [ ] Handoff: `/sdd-develop - <portable-plan-path> - Step 1`
- [ ] Initial progress `0/N`
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall