ワンクリックで
flywheel-drift-check
Check if the codebase has drifted from the implementation plan.
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。
メニュー
Check if the codebase has drifted from the implementation plan.
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
Relentless goal/design interview that sharpens framing and writes durable docs (brainstorm artifact, ADRs, glossary) as decisions crystallize. Use when refining a flywheel goal, pressure-testing scope, or "grill with docs".
Start or resume the full agentic coding flywheel. Drives the complete workflow: scan → discover → plan → implement → review.
Set up flywheel prerequisites for this project.
One-shot diagnostic of every flywheel dependency — MCP connectivity, Agent Mail liveness, br/bv/ntm/cm binaries, node version, git status, dist-drift, orphaned worktrees, and checkpoint validity. Use when debugging toolchain issues, before starting a new session, after /flywheel-cleanup, or as a CI gate.
Strategic gap analysis between vision (AGENTS.md / README.md / plan docs) and what's actually implemented. Converts gaps into beads, optionally launches a swarm. Use when "reality check", "where are we really", "gap analysis", "did we drift", or before declaring a long-running project done.
Launch a parallel swarm of agents to implement multiple beads simultaneously.
| name | flywheel-drift-check |
| description | Check if the codebase has drifted from the implementation plan. |
Run a strategic drift check. $ARGUMENTS
Read the plan document path from .pi-flywheel/checkpoint.json (planDocument field). If not found, scan docs/plans/ for the most recent plan file.
Read the current bead statuses: br list --json via Bash.
Run bv --json via Bash for graph analysis (cycles, orphans, bottlenecks).
Use Agent(Explore) to compare current code state against the plan:
Display a drift report:
✅ On track: N beads
⚠️ Potentially stale: N beads
❌ Blocked: N beads (dependency issues)
🔄 New opportunities: (not in original plan)
If drift is significant (≥3 stale or new-opportunity beads), surface a follow-up via AskUserQuestion:
AskUserQuestion(questions: [{
question: "Significant drift detected (<N> stale + <M> new-opportunity beads). What do you want to do?",
header: "Drift",
options: [
{ label: "Polish-loop the plan", description: "Call flywheel_approve_beads(action: 'polish') and refine bead graph in place (Recommended for tactical drift)" },
{ label: "Run full reality-check", description: "Drift-check is the lightweight version; reality-check is the deep strategic pass — invoke /agent-flywheel:flywheel-reality-check (or read skills/start/_reality_check.md). Best when drift suggests a vision-vs-implementation gap, not just bead-graph staleness." },
{ label: "Ignore — log and continue", description: "Acknowledge drift but proceed without changes" }
],
multiSelect: false
}])
Route on the answer:
flywheel_approve_beads with action: "polish" via the agent-flywheel MCP server.skills/start/_reality_check.md and execute its depth-selection flow.For minor drift (<3 affected beads), skip the follow-up and just ask the original "polish?" question — the lightweight path is sufficient.