スキル executor-brainstorm

📦

executor-brainstorm

Name: executor-brainstorm
Author: atri10

コンテンツリビジョン r1 中リスク ⚙️ 外部コマンド📁 ファイルシステムへのアクセス

Compare Feature Designs Before Specification and Planning

Rough feature ideas can become specifications before alternatives and failure cases are examined. This skill structures research, independent concepts, critique, and human decisions into a recorded design.

対応: Claude Codex Code(CC)

📊 69 十分

Maintain this Skill? Claim it 📊 69 十分

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 "executor-brainstorm" from https://skillstore.io/skills/atri10-executor-brainstorm.md and its manifest at https://skillstore.io/api/skills/atri10-executor-brainstorm/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/atri10-executor-brainstorm.md 署名済み manifest GET /api/skills/atri10-executor-brainstorm/manifest 署名済み lockfile GET /api/skills/atri10-executor-brainstorm/lockfile

テストする

「executor-brainstorm」を使用しています。 Illustrative request: compare self-service tenant onboarding approaches with low operational cost.

期待される結果:

Brief: let tenant administrators onboard without operator assistance; agree success measures and constraints before concept exploration.
Concepts: manual approval with automation, guided self-service, and asynchronous provisioning, plus the current process as a baseline.
Stress test: examine duplicate submissions, failed provisioning, authorization mistakes, and recovery paths.
Decision: the human chooses guided self-service and records the operational tradeoff.
Handoff: requirement candidates for each flow, an initial delivery slice, and unresolved assumptions.

「executor-brainstorm」を使用しています。 Illustrative request: choose how a specification should split into implementation plans.

期待される結果:

Options: split by user journey, by domain boundary, or by delivery increment.
Known approaches: compare each split with existing repository conventions and explain what it borrows.
Adversarial pass: examine integration dependencies, migration order, and rollback costs for the favored split.
Outcome: record the human choice, rationale, and planning phase that receives the decision.

セキュリティ監査

中リスク

v1 • 2026/10/5 バージョン別レポートを開く

Of 132 static findings, 129 are formatting, fixed sibling references, or legitimate local checks; three retain uncertainty about required, unreviewed helper execution. One semantic finding identifies missing confidentiality review for public design records. No evidence found of malicious prompt injection, credential exfiltration, or network reconnaissance in the four supplied files.

スキャンされたファイル

1,204

解析済み行数

レビュー項目

誤検知を無視

確認済みのセキュリティ上の懸念 (1)

中

SKILL.md:75-90 SKILL.md:275-281 prior-art-scout-prompt.md:138-157

Public Records Lack Confidentiality Review

The workflow records business constraints and repository findings in a store described as public. Only a secrets scan is required, leaving personal and proprietary information without explicit publication approval or redaction.

The source explicitly calls the store public and requires recording business constraints and repository details. Disclosure depends on actual inputs; no transmission is demonstrated.

機能レビュー項目 (3)

これらは、このスキルに期待される可能性のある実際のローカル機能であるため、レビューが必要ですが、確認済みの悪意ある動作としてはカウントされません。

中

SKILL.md:41-44

Ruby/shell backtick execution

```bash

The Bash example invokes an external exec-id helper absent from the reviewed package. Its effects cannot be verified, so execution remains a dependency risk.

中

SKILL.md:267-269

Ruby/shell backtick execution

1. `../executor/scripts/exec-store-check` — no B2 or B3 finding for this

The workflow requires running exec-store-check from a sibling package absent from this review. Unknown helper effects remain an execution risk.

中

SKILL.md:275-277

Ruby/shell backtick execution

4. `../executor/scripts/exec-scan-secrets <session dir>` — exit 0; the

The workflow requires running exec-scan-secrets on session files, but its implementation is outside this review. Its access and side effects remain unverified.

リスク要因

⚙️ 外部コマンド (50)

📁 ファイルシステムへのアクセス (6)

SKILL.md:43 SKILL.md:50 SKILL.md:219 SKILL.md:231 SKILL.md:267 SKILL.md:275

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/atri10-executor-brainstorm/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/atri10-executor-brainstorm/security.svg)](https://skillstore.io/skills/atri10-executor-brainstorm?utm_source=security_passport_badge)

HTMLバッジ

<a href="https://skillstore.io/skills/atri10-executor-brainstorm?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/atri10-executor-brainstorm/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/atri10-executor-brainstorm.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

atri10. (2026). executor-brainstorm security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/atri10-executor-brainstorm/audits/1

BibTeX形式の引用

@techreport{atri10-atri10-executor-brainstorm-2026,
  author = {atri10},
  title = {executor-brainstorm security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/atri10-executor-brainstorm/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: "executor-brainstorm security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "atri10"
date-released: "2026-10-05"
url: "https://skillstore.io/skills/atri10-executor-brainstorm/audits/1"
identifiers:
  - type: other
    value: "skillstore:atri10-executor-brainstorm: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

Evaluate a New Product Capability

Compare distinct feature concepts against agreed user outcomes, costs, and constraints before selecting a design.

Resolve an Architecture Question

Generate at least three options for storage or synchronization, examine precedent and failure cases, and record the chosen approach.

Prepare Planning Boundaries

Use decision mode to compare specification decomposition options and record the selected plan boundaries before planning begins.

これらのプロンプトを試す

Frame a Rough Idea

Use executor-brainstorm for this idea: [idea]. Start with a design brief and ask me to approve actors, outcomes, constraints, and evaluation weights.

Compare Feature Concepts

Design [feature] in [repository]. Research [focus areas], compare three independent lenses and the status quo, critique leading concepts, and ask for my decision.

Resolve One Design Question

Run decision mode for [question]. Generate at least three options, identify known approaches, challenge the favorite, and record my choice for [phase].

Prepare a Constrained Design Handoff

Run design mode for [capability] under [constraints]. Compare robustness, experience, and reuse lenses; test failure flows, rollback, and tenfold scale. Document requirement candidates and increments. Obtain my approval before deciding, and review confidentiality before writing public records. Report unavailable helpers instead of claiming verification.

ベストプラクティス

Agree measurable outcomes, constraints, and evaluation weights before generating concepts.
Keep explorers independent and record every critique finding with an explicit disposition.
Review public records for confidential information and verify trusted companion helpers before claiming completion.

回避

Choose the winning concept first and adjust evaluation weights afterward.
Let explorers read sibling concepts or let a concept author critique their own work.
Mark a session decided without the human choice or claim verification when required helpers are unavailable.

よくある質問

When should I use design mode?

Use design mode for a new feature or capability that needs a complete design before specification.

When should I use decision mode?

Use decision mode for one open design question or planning decomposition. It compares at least three options and normally runs without subagents.

Who chooses the final design?

The human chooses after reviewing the concepts and critique. If no choice is made, the session remains draft.

What does the skill produce?

Design mode records a dossier, concepts, prior art, critique, risks, and handoff material. Decision mode records options, an adversarial pass, and an outcome.

Are these four files enough for the complete workflow?

No. The complete workflow references sibling Executor helpers and documentation. Parallel design exploration also requires subagent support in the host harness.

Can I include confidential project information?

The instructions describe the store as public. Use private storage or approved, redacted inputs; a secrets scan does not establish publication permission.