AGENT SKILL

apex-review

Reviews a technical proposal from the perspective of organizational decision-makers (CTO, technical lead, QA lead, conference reviewers, etc.) and generates an HTML presentation deck. Outputs approval gate results per persona and connects to sim-conductor for improvement suggestions.

License 许可
MIT
最近更新
2026-11-15
Stars
★ 33
Forks
1

01 前言 Front Matter

name
apex-review
model-note
session-inherit — Sonnet base is first-class (sonnet_floor_doctrine.md); depth-critical judged steps route to dispatch (opus agent / cross-family sidecar, consent-gated), never a substrate requirement
description
Reviews a technical proposal from the perspective of organizational decision-makers (CTO, technical lead, QA lead, conference reviewers, etc.) and generates an HTML presentation deck. Outputs approval gate results per persona and connects to sim-conductor for improvement suggestions.
allowed-tools
[ "Read", "Write", "Bash", "Agent" ]
user-invocable
true

03 SKILL.md 正文

apex-review — Decision-Maker Review Layer

Reviews proposals from the perspective of top-level organizational decision-makers, not field practitioners. Technically excellent and "this person will approve it" are different things.

Role distinction from hub-persona-auditor

  • hub-persona-auditor: Validates reader comprehension of a finished asset (reading experience)
  • apex-review: Evaluates decision-maker approval likelihood of a technical proposal (decision-making experience)

Triggers

  • /apex-review
  • "Will this work for the decision-maker?", "Will the CTO approve this?", "What do I need to present to the team lead?"
  • "Review from an executive perspective", "Generate presentation slides" (when decision-maker audience is specified)
  • When agent-composer identifies a task requiring a decision-maker approval gate
  • "Check this before I report to my manager"
  • "Does this look convincing?"
  • "Find objections they might raise in advance"
  • "Review before the executive briefing"
  • "Do you think this will pass?"

Built-in Decision-Maker Personas

Code Role Core Evaluation Criteria Auto-Reject Triggers
CTO Chief Technology Officer Strategic alignment · competitive advantage · tech vision "Everyone else is already doing this" / "Feels like a personal project"
TD Technical Lead / Team Lead Implementation risk · team capability · operational burden · cost Vague timelines / conflicts with existing stack
QAL QA Lead / Quality Owner Quality standards · testability · rollback plan · compliance No validation plan / failure scenarios not documented
CONF Conference Organizer / External Reviewer Novelty · demo-readiness · audience value · reproducibility Overloaded with internal jargon / demo not possible

If the user does not specify a target persona, it will be confirmed in Step 0.


Step 0. Receive Topic + Target

If no input is provided:

Please provide the apex-review target.

  Proposal summary: (What are you seeking approval for?)
  Target decision-maker: [CTO / TD (Technical Lead) / QAL (QA Lead) / CONF (Conference)] — multiple allowed
  Proposal file: (provide path or paste content if available)
  Output format: [HTML presentation deck (recommended) / Markdown report]

If no proposal file is provided, collect content through conversation.


Step 1. Structure the Proposal

Organize the received content into the following slide structure.

[S1] One-line thesis      — "What needs to be done and why"
[S2] Current state / problem — What is not working right now
[S3] Proposal             — What changes and how
[S4] Expected outcomes    — Quantified or verifiable change
[S5] Implementation plan  — Phases, timeline, ownership
[S6] Risks + mitigations  — Failure scenarios + rollback
[S7] Requests             — What is needed from the decision-maker

If any slide is missing during structuring, request the user to fill it in.


Step 2. Generate HTML Presentation Deck

Generate as apex_review_deck_YYYYMMDD.html.

HTML structure rules:

  • Each slide = <section class="slide"> block
  • Per slide: title + body + (if applicable) decision-maker attention note
  • Navigation: keyboard ←→ or buttons
  • Style: clean dark background / large headlines / emphasized key figures
<!-- Slide example -->
<section class="slide">
  <h2>Current State · Problem</h2>
  <p class="body">…</p>
  <aside class="exec-note CTO">CTO attention: …</aside>
</section>

Notify the user of the file path after saving.


Step 3. Persona Review

Evaluate the proposal for each target decision-maker.

Output format:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
apex-review — Decision-Maker Review Results
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

[CTO Review]
  Verdict: ✅ Approved / ⚠️ Conditional / ❌ Rejected
  Rationale: …
  Conditions (if conditional): …
  Critical weaknesses: …

[TD Review]
  Verdict: …
  …

Step 4. Approval Gate

All ✅ Approved: Gate passed → proceed to Step 5 1+ ⚠️ Conditional: Output condition list → request user decision 1+ ❌ Rejected: Output rejection rationale + improvement directions → suggest revision and re-review

After gate verdict:

Gate result: {Passed / Conditionally passed / Rejected}

Choose your next step:
  A. Revise and re-review
  B. Proceed to sim-conductor in current state (deep validation)
     ← strongly recommended when verdict is "Conditionally passed"
  C. Use HTML deck only (as reference)
  D. Exit

"Conditionally passed" → option B is the default path: Conditions listed by personas remain unresolved until sim-conductor Area E runs challenger/beginner/main-player validation against them. Choosing C or D without sim-conductor leaves those conditions unverified — present B as the default choice and require explicit opt-out.


Step 5. Deep Validation (Optional)

When option B is selected:

Connect to sim-conductor Area E.
  Target: apex_review_deck_YYYYMMDD.html
  Evaluation criteria: Decision-maker perspective (CTO / TD / QAL / CONF personas)
  Output: Improvement recommendations by M/S/R tier

Incorporate sim-conductor results into the HTML deck and save the revised version.


Operating Notes

  • Invocable without input: Content collected through conversation in Step 0
  • Customizable personas: If the user specifies "our executives prioritize X", it will be reflected
  • Iterative review: After revision, re-run from Step 3 (HTML decks distinguished by version date)
  • Standalone or within agent-composer: In complex tasks, agent-composer will automatically include this skill

Done When

All steps 0–4 completed
+ Verdict (✅/⚠️/❌) output for all target decision-makers
+ Gate verdict (Passed / Conditionally passed / Rejected) stated
+ Deck artifact exists on disk: `apex_review_deck_YYYYMMDD.html` (Step 2) — confirm by printing its
  path and byte size; this is the skill's only durable output, so "reviewed but no deck" is NOT done
+ If the gate verdict is `Conditionally passed`: the `/sim-conductor Area E` verdict has been
  RECEIVED and folded in (the mandatory-next gate below). Without that verdict the run is not done,
  regardless of the A/B/C/D selection
+ User next step selection (A/B/C/D) confirmed

→ Mandatory next (Conditionally passed verdict): /sim-conductor Area E — run against apex_review_deck_YYYYMMDD.html before the proposal is considered ready for submission. Skipping sim-conductor on a conditional pass means unresolved persona conditions go unverified; present option B as default and require the user to explicitly choose A/C/D to override.

Wait for sim-conductor Area E verdict before declaring the proposal submission-ready. (Conditionally passed path: proposal is not done until sim-conductor verdict folds back.)

04 同仓库更多 Skills

chrono-meta
chrono-meta/forge-harness

todo-orchestrator

TODO CLI 프로젝트의 기능 추가·버그 수정·리팩터링·테스트 보강을 개발자와 QA 에이전트로 조율한다. todo 앱 관련 코드 작업 요청에 사용한다. 다시 실행, 재실행, 수정, 보완, 이전 결과 개선, QA만 다시 같은 후속 요청에도 사용한다. 단순 질문에는 사용하지 않는다.

★ 33
chrono-meta
chrono-meta/forge-harness

todo-dev

TODO CLI(todo/cli.py)에 명령을 추가하거나 버그를 고치고 테스트를 작성하는 방법. todo 앱 코드 수정·기능 추가·테스트 작성·리팩터링 작업에 사용한다. 수정, 보완, 다시 구현 요청에도 사용한다.

★ 33
chrono-meta
chrono-meta/forge-harness

todo-dev-team

TODO CLI(todo/cli.py) 개발 작업을 개발자·테스터·리뷰어 팀으로 진행하는 오케스트레이터. 명령 추가, 버그 수정, 테스트 작성, 리팩터링, 커밋·푸시 요청에 사용한다. 다시 실행, 재실행, 수정, 보완, 리뷰 반영, 테스트만 다시 같은 후속 요청에도 사용한다. 단순 질문에는 쓰지 않는다.

★ 33
chrono-meta
chrono-meta/forge-harness

todo-dev

TODO CLI 개발 작업(명령·기능 추가, 버그 수정, 테스트 작성, 리뷰, 커밋·push)을 개발자·테스터·리뷰어 에이전트 팀으로 진행하는 오케스트레이터. todo CLI, todo.json, todo/cli.py 관련 구현·수정·테스트 요청에 사용한다. 재실행, 수정, 보완, 이전 결과 개선, 테스트만 다시, 리뷰만 다시 요청에도 사용한다. 단순 질문에는 쓰지 않는다.

★ 33
chrono-meta
chrono-meta/forge-harness

weekly-summary-write

정리된 회의록으로 주간 회의 요약 문서를 작성하는 형식과 문체 규칙. 주간 요약 작성, 요약 문서 수정·보완, 검토 의견 반영 시 사용한다.

★ 33
chrono-meta
chrono-meta/forge-harness

weekly-meeting-summary

주간 회의록을 정리하고 요약 문서를 만드는 전체 작업을 조율하는 오케스트레이터. 회의록 요약, 주간 회의 정리, 주간 보고용 요약 요청 시 사용한다. 다시 실행, 재실행, 요약 수정·보완, 이전 결과를 바탕으로 한 개선, 정리만/검토만 다시 같은 후속 요청에도 사용한다.

★ 33