メニュー
Plan phase — break an approved spec into an intent-driven task list for OpenClaw crew execution. Verify assumptions with a scout sub-agent, map files to owners, write WHAT-not-HOW tasks with acceptance criteria and Depends on, then hand off to flow:execute.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Choose how to handle a new coding request — native for tiny edits vs a fakoli-claw flow for multi-file, risky, design, or explicit crew/Fakoli work.
Design phase — refine an idea into an approved spec through structured one-question-at-a-time dialogue, propose 2-3 approaches, present the design section by section, write the spec, and hand off to flow:plan. Works headless over any channel.
Execute phase — wave-based crew dispatch on OpenClaw with critic gates and evidence-based verification. Load an intent-driven plan, group tasks into dependency waves, sessions_spawn specialists in parallel per wave (local SGLang tier), run a critic gate after every code wave, then a sentinel evidence gate.
Ship phase — re-run fresh tests, then present exactly four options (merge locally, push+PR, keep as-is, discard) and execute only the explicit choice. Never auto-merge, auto-push, or discard without the typed word discard.
Fast path — skip brainstorm/plan/waves for small tasks under 3 files. Estimate scope, dispatch a single specialist (welder by default), run verification, run a one-cycle critic gate, done. Escalates to flow:execute if scope or fixes exceed the limit.
Verify phase — evidence-based validation. Detect language, run the full check suite, dispatch the sentinel against the plan's acceptance criteria, and produce a pass/fail scorecard where every PASS cites fresh command output from this session.
| name | flow-plan |
| description | Plan phase — break an approved spec into an intent-driven task list for OpenClaw crew execution. Verify assumptions with a scout sub-agent, map files to owners, write WHAT-not-HOW tasks with acceptance criteria and Depends on, then hand off to flow:execute. |
Read an approved spec, verify assumptions with a scout, write an intent-driven task list, and
hand off to flow:execute. Plans describe WHAT to achieve — never HOW.
openclaw agents list | grep fakoli-scout), spawn:
sessions_spawn(agentId="fakoli-scout", task=…) to verify libraries/APIs exist, find existing
patterns, and flag partial implementations. Derive a plan- run id and inject the absolute
status path. sessions_yield, read findings before writing tasks. If scout finds a broken
assumption (missing lib/API), flag it to the user before proceeding.docs/plans/<YYYY-MM-DD>-<feature>.md (or the path CLAUDE.md specifies).flow:execute with it.| Work | Agent |
|---|---|
| New modules / interfaces / types | fakoli-guido |
| Wiring new code into existing systems, integration | fakoli-welder |
| Research / codebase exploration / lib verification | fakoli-scout |
| Plugin manifests / commands / structure | fakoli-smith |
| README / docs / changelogs | fakoli-herald |
| Infra / CI / config | fakoli-keeper |
| Code review (review-only task) | fakoli-critic |
| Test/acceptance validation (review-only task) | fakoli-sentinel |
| Any role when crew absent | main / general |
Tier note: assignment is also a cost lever — welder/smith/scout/herald/keeper run on the local SGLang tier (free, parallel); guido/critic/sentinel on GPT-5.5. Prefer local specialists for high-fan-out work.
# <Feature> — Execution Plan
**Goal:** one sentence.
**Spec:** docs/specs/<date>-<topic>.md
**Language:** TypeScript | Python | Rust (detected)
**Crew:** fakoli-claw crew (N agents) | generic (crew not installed)
### Task 1: <name>
**Intent:** verb-first outcome, one sentence.
**Acceptance criteria:** 2–5 independently checkable bullets.
**Scope:** exact/file/paths
**Agent:** fakoli-guido | fakoli-welder | … | generic
**Verify:** exact command proving done
**Depends on:** (none) | Task N
Crew detection command: openclaw agents list 2>/dev/null | grep -c '^- fakoli-' (0 = generic mode).