scoutlisted
Install: claude install-skill aimsise/simple-workflow
Investigate and plan: $ARGUMENTS
Invocation policy: Do not auto-invoke. Only invoke when explicitly called by name by the user (e.g. `/scout <topic>` or `/scout` on an active ticket) or by another skill via the Skill tool. In practice, `/autopilot` Step 3b chain-calls `/scout` once per ticket as part of the per-ticket `scout` → `impl` → `ship` pipeline. `disable-model-invocation: false` is intentional because the `/autopilot` chain-call uses the Skill tool, which would not resolve if the flag were flipped to `true`; flipping it would break the chain-call surface for `/autopilot` while leaving direct user invocation (`/scout <topic>` or `/scout` on an active ticket) superficially intact.
## Mandatory Skill Invocations
The following skill invocations are **contractual** — `/scout` MUST delegate to each of these via the Skill tool. `/scout` is a thin orchestrator and performs no research/planning work itself; its entire purpose is to chain /investigate and /plan2doc with ticket-aware arguments. Any bypass is a contract violation and will be detected by the skill invocation audit (Phase A+).
| Invocation Target | When | Skip consequence |
|---|---|---|
| `/investigate` (Skill) | Step 3 — always, after ticket resolution | No `investigation.md` written to the ticket dir; downstream `/plan2doc` has no research context, producing a weaker plan. `/autopilot`'s post-scout artifact verification triggers `[PIPELINE] scout: ARTIFACT-MISSING — investigation.md` |
| `/plan2doc` (Skill)