Menü
Use when start point is unclear (idea, repo, bug, feature) before plan/build. Not when route already clear.
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
Use when a plan or concrete tasks exist and user said build/go/implement. Not to invent scope or claim done without verify.
Use when goal/spec is clear and you need ordered tasks before writing code. Not while still clarifying.
Use when about to claim done/works/fixed, or user asks test/check/prove. Run evidence; not for pure planning chat.
Use when another installed plugin/skill may do the job better, or to route specialists. ACC stays workflow owner.
Use when user wants an explicit recorded decision or blocker saved to artifacts. Not silent monitoring.
Use when request is too vague to plan or build. Not when goal and constraints are already clear.
| name | onboard |
| description | Use when start point is unclear (idea, repo, bug, feature) before plan/build. Not when route already clear. |
Reply rule:
Use $onboard when the plugin needs to understand the starting point before planning or building.
Step 1 first: run /hooks, trust anyone-can-code, restart Codex. That is
auto-memory. Without it ACC forgets between sessions. Repeat after plugin
updates. Then $setup if project folders are missing.
idea: a vague or partial idea, likely routes to $clarify.written-spec: requirements already exist, likely routes to $plan.existing-repo: repo scan first, then route into fast path or full path.feature-request: concrete change in an existing codebase, usually fast path.bug-fix: bug or failure, route into repair and verification.polish-review: optimization or review pass.ship-verify: closing and verification path.requirement-change: update plan/state, then resume from the last valid step..codex/anyone-can-code/state/workflow.jsonAGENTS.mdDo not ask broad setup questions if the repo and the user request already answer them.
Use scripts/front_door.py as the shared classifier. Do not create a second
entry-mode classifier in this skill.
Memory onboarding shows selected Markdown storage and optional viewer mode. Never inspect or import existing session files unless user selected exact paths, scope, and confirmation.
Next: $clarify if still vague, else $plan or $execute / $fix by entry mode.
When this skill's job is finished: