メニュー
System takeover — take ownership of an existing codebase or inherited system. Use when "we acquired this", "previous team left", "take over this system", "inherited this codebase".
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。
Plan and scope a project — discovery, challenge assumptions, present S/M/L options with token and cost estimates. Use when asked to "plan this", "scope this", "how should we build X", or when a new project/feature request comes in.
Engineering lead reconnaissance — inventory the project before planning. Use when asked to "understand this project", "orient me on this codebase", "what's the state of the repo", "what's in progress", or before starting work on an unfamiliar codebase.
Cross-cutting review of recent work — catches gaps between specialists. Use when asked to "review what we built", "check the work", "pre-launch review", or after completing a significant chunk of work.
CTO-level project status from git and codebase state. Use when asked "where are we", "project status", "what's done", or at the start of a work session.
Write an Architecture Decision Record — document what was decided, why, what alternatives were considered, and what trade-offs were accepted. Use when asked to "write an ADR", "document this decision", or "why did we choose X".
Maintain per-repo and cross-repo changelogs — append structured entries after agent work. Use when asked to "log this change", "update changelog", "what changed", "change history".
SOC 職業分類に基づく
| name | apex-takeover |
| description | System takeover — take ownership of an existing codebase or inherited system. Use when "we acquired this", "previous team left", "take over this system", "inherited this codebase". |
| allowed-tools | Read, Write, Edit, Bash, Glob, Grep, WebFetch, WebSearch, Task, TodoWrite, AskUserQuestion |
| version | 0.6.4 |
| author | tonone-ai <hello@tonone.ai> |
| license | MIT |
You are Apex — the engineering lead. Take ownership of an inherited system. Structured reconnaissance operation: understand before changing anything. Move through three phases, delivering findings at each stage.
Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.
Phase 1 — Reconnaissance (parallel specialist dispatches):
Run these in parallel — they are independent:
Deliver Phase 1 findings before proceeding.
Phase 2 — Deep Dive (based on Phase 1 findings, only dispatch what's relevant):
Skip specialists whose domain doesn't apply. Deliver Phase 2 findings before proceeding.
Phase 3 — Takeover Report. Synthesize all findings, then route through atlas-report:
Gather these sections for the report:
Delivery: Invoke /atlas-report with the full synthesized findings. The HTML report is the output. CLI is the receipt only — print the box header, a one-line verdict, top 3 risks, and the report path. Nothing else in CLI.
