📦

council

Name: council
Author: warpdotdev

コンテンツリビジョン r1 安全 ⚙️ 外部コマンド

Compare Decisions with an AI Council

Complex decisions can suffer from narrow analysis and hidden assumptions. This skill coordinates diverse agents, compares their evidence, and produces one actionable recommendation.

対応: Claude Codex Code(CC)

🥉 78 ブロンズ

Maintain this Skill? Claim it 🥉 78 ブロンズ

自分のエージェントでインストール

このリクエストをエージェントにコピーしてください。正規の Skill ページとマニフェストが含まれています。

エージェントリクエスト

Review the Skillstore skill "council" from https://skillstore.io/skills/warpdotdev-council.md and its manifest at https://skillstore.io/api/skills/warpdotdev-council/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

エージェントは引き続き計画を提示し、セキュリティポリシーで必要な確認を求める必要があります。

エージェントが読めるリソース

AI エージェント、クローラー、スクリプトがページ全体ではなく整理されたコンテキストを必要とする場合は、これらのリンクを使ってください。

Markdown詳細 GET /skills/warpdotdev-council.md 署名済み manifest GET /api/skills/warpdotdev-council/manifest 署名済み lockfile GET /api/skills/warpdotdev-council/lockfile

テストする

「council」を使用しています。 Should we keep the current cache or adopt the proposed distributed cache?

期待される結果:

Recommendation: keep the current cache for this release. The proposal improves scale, but migration risk and weak failure testing outweigh the immediate benefit.

「council」を使用しています。 Review this authentication fix from correctness, security, and rollout perspectives.

期待される結果:

Consensus: the fix closes the reported bypass.
Disagreement: reviewers differ on session invalidation scope.
Next step: add expiration tests and deploy behind staged monitoring.

「council」を使用しています。 Decide whether to refactor now or stabilize the current implementation.

期待される結果:

Final call: stabilize first, measure the remaining defects, and schedule the refactor only if the measured maintenance cost justifies it.

セキュリティ監査

安全

v1 • 2026/9/4 バージョン別レポートを開く

All six static findings are false positives caused by Markdown backticks, a fenced example, and ordinary model-selection prose. The skill contains no executable code, system reconnaissance, prompt injection, or semantic security concerns.

スキャンされたファイル

131

解析済み行数

レビュー項目

誤検知を無視

リスク要因

⚙️ 外部コマンド (4)

SKILL.md:44 SKILL.md:46 SKILL.md:52 SKILL.md:104-123

最新の完了済み静的・セマンティック監査では、確認済みのセキュリティ検出事項は見つかりませんでした。これは、スキルに副作用がないことを証明するものではありません。

監査者: codex

このレポートを共有・引用

バージョン付き評価レポート、中立的なバッジ、埋め込みカード、引用を共有できます。Skillstore は証拠を報告しますが、この Skill が安全かどうかは判断しません。

バージョン別レポートを開く

レポートリンクをコピー

https://skillstore.io/skills/warpdotdev-council/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdownバッジ

[![Skillstore security assessment](https://skillstore.io/badges/skills/warpdotdev-council/security.svg)](https://skillstore.io/skills/warpdotdev-council?utm_source=security_passport_badge)

HTMLバッジ

<a href="https://skillstore.io/skills/warpdotdev-council?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/warpdotdev-council/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/warpdotdev-council.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

学術引用 (APA · BibTeX · CFF)

APA形式の引用

warpdotdev. (2026). council security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/warpdotdev-council/audits/1

BibTeX形式の引用

@techreport{warpdotdev-warpdotdev-council-2026,
  author = {warpdotdev},
  title = {council security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/warpdotdev-council/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "council security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "warpdotdev"
date-released: "2026-09-04"
url: "https://skillstore.io/skills/warpdotdev-council/audits/1"
identifiers:
  - type: other
    value: "skillstore:warpdotdev-council:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore スコア

このスコアの理由証拠の信頼度: 中

アーキテクチャ

保守性

コンテンツ

コミュニティ

仕様準拠

作成できるもの

Review an architecture choice

Compare competing designs for correctness, implementation cost, operational risk, and testability before committing to one approach.

Challenge a risky code change

Ask independent reviewers to inspect a proposed fix for regressions, security concerns, and missing tests.

Assess an incident response

Combine implementation, reliability, and contrarian perspectives into a prioritized recovery and prevention recommendation.

これらのプロンプトを試す

Get a second opinion

Run a council on [question]. Compare [option A] and [option B], then recommend one using [decision criteria].

Review a technical proposal

Form a read-only council to review [proposal or artifact]. Assess correctness, implementation effort, testing needs, and operational risks.

Red-team a planned change

Use diverse models to evaluate [change]. Assign architecture, implementation, and adversarial perspectives. Cite evidence and identify unresolved assumptions.

Produce an evidence-weighted decision

Investigate [decision] with a model-diverse council. Compare findings by evidence quality, resolve disagreements, and separate immediate actions from optional hardening.

ベストプラクティス

Define one precise decision, relevant artifacts, constraints, and evaluation criteria before launching the council.
Use distinct models and perspectives, then judge reports by cited evidence instead of vote count.
Keep read-only agents from editing files and isolate implementation agents in separate worktrees.

回避

Do not launch several agents with identical models, prompts, and analytical roles.
Do not treat lifecycle completion as proof that a report contains useful evidence.
Do not present raw agent transcripts when the user needs a concise recommendation.

よくある質問

What problems suit this skill?

It works best for architecture choices, risky fixes, code reviews, rollout decisions, incident analysis, and other judgment-heavy questions.

Does every council require three agents?

No. Three diverse members are the preferred default, but availability, scope, and user constraints can justify another size.

Can council members edit code?

Yes, when requested. Each local implementation agent should use an isolated worktree and branch to prevent conflicts.

What happens when model diversity is unavailable?

The skill uses the closest supported models and preserves diversity through distinct specialist or adversarial perspectives.

How are conflicting recommendations resolved?

The final synthesis weighs evidence quality, decision criteria, risks, and unknowns instead of choosing the majority view.

Does the skill guarantee a correct decision?

No. Its recommendation depends on available evidence, agent capabilities, and the accuracy of the supplied context.

開発者情報

作成者

warpdotdev

ライセンス

MIT

Skillstore リビジョン

バージョンに関する注意

作者はバージョンを宣言していません。

リポジトリ

https://github.com/warpdotdev/common-skills/tree/b811c24365ae505bfc9646458957b886e29110b5/.agents/skills/council

参照

096c7daa4d0d974771dad12d81c3d0d4688617b9

メンテナンスの新しさ

2026/9/8

利用状況

1 ダウンロード · 0 閲覧

ファイル構成

📄 SKILL.md