tibursocampos/agent-dev-toolkit / core/skills/orchestrate-analyze
orchestrate-analyze
VerifiedOrchestrated Delivery O1: triage a feature, spawn conditional Agent specialists, write FEATURE.md + CONTINUITY + US/TS under features/NNN-slug/. No app code. Use when invoking /orchestrate-analyze.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 18.5 KB |
| reference.md | markdown | 1.8 KB |
| references/arch-confirm.md | markdown | 4.0 KB |
| references/boundaries-handoff.md | markdown | 3.5 KB |
| references/command.md | markdown | 3.5 KB |
| references/intent-classification.md | markdown | 5.2 KB |
| references/memory-bank-gate.md | markdown | 881 B |
| references/must-not.md | markdown | 4.4 KB |
| references/process-common.md | markdown | 3.8 KB |
| references/spawn-map.md | markdown | 3.7 KB |
| references/story-synthesis.md | markdown | 18.8 KB |
| references/triage.md | markdown | 5.0 KB |
references/must-not.md
View on GitHub## Must not (full)
- Skip Step 0 Memory Bank Gate (unless explicit user `skip-memory-bank`)
- Create or place `memory-bank/` under `features/NNN-slug/`
- Duplicate memory-bank body into CONTINUITY (path + status only)
- Dump entire memory-bank into the parent orchestrator context
- Write application/production code or tests (`*.cs`, `*.tsx`, `*.ts`, `*.js`, `*.vue`, `*.py`, migrations, etc.)
- Show the current branch or ask for a branch switch. Write analysis artifacts on the current branch, including `main`, `master`, and `develop`
- Call `*-developer` / `developer` to **implement** code (suggesting the trivial shortcut is allowed)
- Skip human backlog approval, or treat silence, an ambiguous or unrelated reply, or a lone emoji as approval. A clear affirmative reply in any language is approval (`{{TOOLKIT_ROOT}}/skills/_shared/agents/QUESTIONS.md`)
- Skip the architecture confirm gate on greenfield / `needs_domain` (no established style), or write final ARCH / pick a silent style default before operator **sim**
- Skip `architect` / `database` (or skip writing mirror ARCH) on brownfield — skip **style re-pick / confirm gate** only
- Leave `memory-bank/architecture.md` draft / `needs-confirm` after ARCH **sim** (point-promote that file; not a full inventory refresh)
- Pile unresolved product/schema choices only in CONTINUITY when `needs_database` / `needs_domain` (write them under `ANALYSIS/` / `ARCH/`; CONTINUITY may pointer only)
- Approve backlog when required `ANALYSIS/` / `ARCH/` / `SEC/` folders are missing for true flags / brownfield
- Present human backlog approval when product artifact quality gates fail (FEATURE missing Problem/Goals/Non-goals — TE01; task-shaped/layer-only candidate promoted to US/TS — TE02; >4 US/TS without explicit FEATURE split rationale — RN03)
- Create `USnn`/`TSnn` for a title that is verb+file/class/script or layer-only (keep as PLAN/refine step; record anti-task-shatter rationale)
- Treat pointer-only cited `.md` as promoted; treat host plan files as O3 input
- Write PRD/PLAN (O2 owns that via `sdd-spec` / `sdd-plan` contracts)
- Create a story folder (`USnn`, `TSnn`, or Bug) while `FEATURE.md` still has an open question (`open_question`)
- Write `PRD/` or `PLAN/` during O1
- Create external work-item tracker or org-only compliance content
- Create ~40 agent files or expand the roster beyond `ROSTER.md`
- Modify toolkit `.gitignore` as part of porting this skill into the toolkit repo; at runtime follow `STORAGE.md` only for consumer repo SDD patterns
- Change the `sdd-develop` one-step-per-session contract
- Create `REFINE/` / `ANALYSIS/` / `ARCH/` / `SEC/` / `PRD/` / `PLAN/` at **repo root**
- Resolve feature paths outside `$Cwd/features/` or `<classic.path>/features/`, or accept `..` segments
- Pass Agent `model` without `SUBAGENT-MODEL.md` gate + user **sim** (or user-named slug); ask model on routine spawns
- Hard-fail when `subagents` is `none` or Agent is unavailable (use **fallback** in-parent notes per `SPAWN.md`; until step 9, keep those notes in the parent — do not create `USnn`, `TSnn`, Bug, `ANALYSIS/`, `ARCH/`, or `SEC/` before the open-question gate; never skip the notes)
- Exceed orchestrate ≤4 concurrent Agent calls without user-approved wave/série (`SPAWN.md`)
- Paste guideline packs into Agent child prompts
- Write SDD artifacts containing OS absolute paths matching `^[A-Za-z]:/` or user-home InstallRoot embeds (`…/.cursor/sdd/…`, `…/.claude/sdd/…`) — use portable paths per `STORAGE.md` § Portable path
- Read full `reference.md` when a `references/<section>.md` exists for the current Process step (`SKILL-REFERENCE-RETRIEVAL.md`)
- Dump integral `memory-bank/` or full PRD into CONTINUITY, child prompts, or FEATURE/STORY (`SR-NO-FULL-DUMP`)
- Omit `## Related` on FEATURE / CONTINUITY / STORY Writes, use `## See also`, stub absent siblings (incl. PRD/PLAN) only for links, or invent `feature-refinement.md` as navigation SoT (`STORAGE.md` § Navigation block / REQ-009)
---
## Explicit exclusions
Do **not** introduce:
- Write of application or test source as part of O1
- `memory-bank/` under `features/NNN-slug/`
- Full memory-bank dump into CONTINUITY or parent chat
- Skip of Step 0 without explicit `skip-memory-bank`
- External work-item tracker CLI/API commands
- External work-item tracker or org-only compliance fields
- Forty aspirational agent files from design.md
- Changes to `sdd-develop` one-PLAN-step-per-session contract
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 471k | Updated | Claude CodeCodex | Verified | Install |
| 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: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |