スキル socratic-story-cartographer
📦

socratic-story-cartographer

v2.2 コンテンツリビジョン r1 安全 🌐 ネットワークアクセス📁 ファイルシステムへのアクセス⚙️ 外部コマンド

Diagnose Story Problems Before Revising

AI story feedback can repeat an unsupported diagnosis and encourage unnecessary changes. This skill tests competing explanations against textual evidence before recommending focused revisions.

対応: Claude Codex Code(CC)
Maintain this Skill? Claim it 🥉 77 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 "socratic-story-cartographer" from https://skillstore.io/skills/oakcoderx-socratic-story-cartographer.md and its manifest at https://skillstore.io/api/skills/oakcoderx-socratic-story-cartographer/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。

テストする

「socratic-story-cartographer」を使用しています。 Illustrative scene: A witness hides a letter, explains its contents to herself, then gives it to the detective without hesitation.

期待される結果:

  • Current belief: The handover lacks a consequential choice rather than an explanation of motive.
  • Evidence anchors: She initially hides the letter, but later gives it away without resistance or a new event.
  • Competing diagnosis: The explanatory passage may remove uncertainty before the handover.
  • Counterfactual test: Removing the explanation leaves the same unmotivated handover, so exposition alone does not explain the problem.
  • Minimal intervention: Establish one immediate cost that makes her decision to hand over the letter meaningful.
  • Protected quality: Keep her motives partly ambiguous rather than adding a complete explanation.

「socratic-story-cartographer」を使用しています。 Illustrative outline: Episode 2 contains an unsupported agreement that causes an episode 6 betrayal. Episode 5 and the ending are locked.

期待される結果:

  • Decision page: Recommend a structural revision proposal for human review, not production approval.
  • Issue route: The unsupported agreement begins in episode 2; its visible consequence is the betrayal in episode 6.
  • Revision range: Inspect episodes 2 through 6 instead of patching only the betrayal.
  • Task objective: Make the agreement depend on an observable pressure or choice.
  • Acceptance question: Does the revised agreement make the later betrayal causally understandable?
  • Stop line: Preserve episode 5 and the locked ending; do not introduce a new subplot.
  • Coverage limit: This brief supports a provisional route, not a complete season assessment.

セキュリティ監査

安全

All 50 static findings are false positives involving narrative terminology, Markdown formatting, installation documentation, a local reference link, and readable Chinese text. No evidence found of malicious prompt injection, data exfiltration, or hidden execution in the reviewed files. Installation examples are optional setup commands; promotion logs describe maintainer activity, not authorization for story reviewers to publish externally.

14
スキャンされたファイル
2,277
解析済み行数
0
レビュー項目
0
誤検知を無視
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
セキュリティ評価

レポートリンクをコピー

https://skillstore.io/skills/oakcoderx-socratic-story-cartographer/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/oakcoderx-socratic-story-cartographer/security.svg)](https://skillstore.io/skills/oakcoderx-socratic-story-cartographer?utm_source=security_passport_badge)

HTMLバッジ

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

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/oakcoderx-socratic-story-cartographer.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

oakcoderx. (2026). socratic-story-cartographer security audit report (audit version 1) [Author version 2.2]. Skillstore. https://skillstore.io/skills/oakcoderx-socratic-story-cartographer/audits/1

BibTeX形式の引用

@techreport{oakcoderx-oakcoderx-socratic-story-cartographer-2026, author = {oakcoderx}, title = {socratic-story-cartographer security audit report (audit version 1)}, institution = {Skillstore}, year = {2026}, number = {1}, url = {https://skillstore.io/skills/oakcoderx-socratic-story-cartographer/audits/1}, note = {Author version 2.2} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "socratic-story-cartographer security audit report (audit version 1)" version: "2.2" type: report authors: - name: "oakcoderx" date-released: "2026-11-09" url: "https://skillstore.io/skills/oakcoderx-socratic-story-cartographer/audits/1" identifiers: - type: other value: "skillstore:oakcoderx-socratic-story-cartographer: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
59
アーキテクチャ
85
保守性
87
コンテンツ
65
コミュニティ
74
仕様準拠

作成できるもの

Review a novel draft

Compare explanations for weak character agency across chapters while preserving voice and ambiguity.

Diagnose a screenplay scene

Check objectives, choices, and state changes before selecting the smallest useful scene revision.

Prepare a season revision handoff

Trace major outline problems to their first appearance and organize supported fixes into prioritized revision tasks.

これらのプロンプトを試す

Beginner: Diagnose one scene
Review the scene below with Socratic Story Cartographer for one loop. Do not rewrite. Give textual evidence and the smallest useful next step. Scene: [text].
Intermediate: Test competing diagnoses
Review this chapter for up to three loops. Compare two plausible causes of the main problem. Test both against textual evidence. Preserve ambiguity and voice. Do not rewrite yet. Chapter: [text].
Advanced: Verify a multi-file character arc
Review these canonical chapter files using the long-work evidence protocol: [files]. Treat [older version] as non-canonical context. Track coverage and test whether the protagonist lacks agency. Inspect later choices before making work-wide claims. Disclose unread material and propose only supported changes.
Expert: Compile a producer review
Review this season outline for up to three loops: [outline]. Apply long-work coverage before producer-outline review. Protect [scenes] and preserve [locked ending]. Do not rewrite or imply approval. Deliver a decision page, issue routes, hard errors, and dependency-ordered revision task cards. Include acceptance questions, stop lines, and the next human approval gate.

ベストプラクティス

  • Provide the relevant canonical material, identify older versions, and state which files belong in the review.
  • Specify your goal, revision permission, protected qualities, and any locked scenes or endings.
  • Ask for competing explanations and evidence before accepting changes; allow early stopping when further loops add no information.

回避

  • Requesting a work-wide diagnosis from isolated excerpts without acknowledging unread material.
  • Treating author intent, model confidence, or benchmark scores as proof that a diagnosis is correct.
  • Requiring repeated rewrites despite weak evidence or treating an AI recommendation as human approval.

よくある質問

Which AI tools can use this skill?
The skill is documented for compatible Agent Skills environments, including Claude Code and Codex. Installation and file access depend on the host environment.
Must every review run three loops?
No. Three loops is the default maximum. The skill should stop earlier when another loop adds no evidence or useful diagnostic change.
Will it rewrite my story automatically?
Revision requires your permission. You can request diagnosis only or allow limited edits while protecting specific story elements.
Can it review long manuscripts or multiple files?
It includes a source manifest and coverage ledger protocol. Conclusions must remain limited when files are inaccessible or reading coverage is incomplete.
Does it guarantee publication or production readiness?
No. It distinguishes text quality from target fit and uncertain acceptance. Human editorial, legal, and production decisions remain separate.
Does story review require external services?
No separate story-review service is specified. The documented installation command uses npm and GitHub; your AI host controls model access and data handling.

開発者情報

作成者

oakcoderx

ライセンス

MIT

作者バージョン

v2.2

Skillstore リビジョン

r1

バージョンに関する注意

作者が宣言したバージョンは有効な SemVer ではありません。

参照

9938f91a9ab0f5763a54332b129295339bf829dd

メンテナンスの新しさ

2026/11/9

利用状況

0 ダウンロード · 0 閲覧

すべて表示