xiduzo/wtf / skills/wtf.setup
wtf.setup
VerifiedThis skill should be used when a user wants to set up WTF in a repository or check that the environment is ready for it — for example "set up wtf", "is wtf ready", "install wtf templates", or "a new dev joined, set them up". Verifies the GitHub CLI and its auth, installs the required gh extensions and the issue and PR templates, configures issue classification, planning, feature scope, and delivery mode, and registers the hooks. Run once per repo when onboarding, or when a contributor joins the project.
shared-references/scope-gates.md
View on GitHub# Vertical-Slice Assessment and Scope Gates
Shared two-stage scope check used by every wtf write-* skill (`wtf.write-epic`, `wtf.write-feature`, `wtf.write-trace`). Also used by `wtf.refine` when scope changes.
The two stages fire at different points. They catch different problems. Both can fire in the same run. Treat them as independent.
## Stage 1 — Vertical-slice assessment (pre-draft)
Runs on **gathered context** (seed idea, research findings, user answers) before any draft exists. Catches structural incoherence when it is cheapest to fix.
The unit must be:
- **Epic** — a coherent, independently deliverable strategic initiative. It produces real user or business value on its own. It is not only a dependency for another epic.
- **Feature** — one step toward the Epic. It carries 1..n co-related user stories that share one Spine. Concrete test: if this feature shipped tomorrow with no other unshipped features, could a domain actor use it and gain business value?
- **Trace** — one pass over the Feature's Spine. It claims exactly one story plus a Scenario Claim: a declared subset of that story's Gherkin scenarios. It runs end-to-end through every layer. It leaves the system releasable when it merges.
Evaluate:
- **Passes** → proceed to draft.
- **Too broad** → propose a split. For an Epic or Feature, propose smaller slices. For a Trace, split by **depth**: a smaller first claim plus Deepening Traces of the same story that re-partition the Scenario Claim. A Skeleton keeps exactly one scenario — the happy path. Never split a Trace by layer. Present the breakdown. Ask the user to confirm before continuing.
- **Has dependencies** → identify explicitly (what this depends on, what depends on this). Record issue numbers for later native-link creation. Do NOT write dependency references into the body yet. The create step handles native links and body writes together.
## Stage 2 — Scope gate (post-draft)
Runs on the **written draft**. Drafting sometimes reveals bundled objectives that were not visible in the abstract. Frame this to the user as a structural check. Do not frame it as a challenge to their earlier answers.
If one or more split signals fire:
1. State which signals you found.
2. Explain the risk in concrete terms — review friction, merge conflict surface, rollback complexity.
3. Propose a concrete split. Give two or three focused candidate titles. Write one sentence each. Use the right shape for the level (for example `[Actor] can [verb] [object]` for Features, or a re-partitioned Scenario Claim for Traces).
4. Ask "How do you want to handle this?" — header `Scope`:
- **Keep the original draft** → proceed with the current draft without splitting
- **Split it** → start over with one of the proposed smaller units
- **Stop here** → exit without creating
On **Split it**, return to the earliest step where the seed was chosen. Carry forward already-gathered research. Only ask again for narrowed-scope clarification.
## Where the per-level signals live
Split signals are level-specific. They belong in the individual skills, not here:
| Level | Signal list |
|---|---|
| Epic | `wtf.write-epic` step 7 — multiple objectives, >8 features, spans contexts without seam, etc. |
| Feature | `wtf.write-feature` step 9 — >6 ACs, stories that do not share the Spine, "and" in capability name, etc. |
| Trace | `wtf.write-trace` — Scenario Claim spans more than one story, claim too large for one agent pass, etc. |
Signals are heuristics, not rigid thresholds. The skills explicitly note this. Use judgment.
A Trace split signal never proposes layer slices (model → API → UI). The escape valve is depth, not layers.
## Using this in refinement
`wtf.refine` reuses both stages when an insight changes scope. Re-run the vertical-slice assessment on the refined intent. Re-run the scope gate on the rewritten sections. Present findings as refinement concerns (not blockers). The user may knowingly accept a broader scope.
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 |