スキル stow
📦

stow

v0.4.2 コンテンツリビジョン r1 安全 ⚙️ 外部コマンド🔑 環境変数🌐 ネットワークアクセス📁 ファイルシステムへのアクセス

Standardize Technical Writing with STOW

Technical documents often lose precision when prose, commands, paths, and structured data are edited inconsistently. STOW applies output contracts and focused prose review while preserving protected literals.

対応: Claude Codex Code(CC)
🥉 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 "stow" from https://skillstore.io/skills/theislampill-stow.md and its manifest at https://skillstore.io/api/skills/theislampill-stow/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/theislampill-stow.md 署名済み manifest GET /api/skills/theislampill-stow/manifest 署名済み lockfile GET /api/skills/theislampill-stow/lockfile

テストする

「stow」を使用しています。 Rewrite a deployment README while preserving its commands and configuration names.

期待される結果:

A concise README with the installation path first, direct headings, consistent terminology, and unchanged commands and configuration names.

「stow」を使用しています。 Review a migration runbook for operational completeness.

期待される結果:

  • Add a precondition stating the required backup and access checks.
  • Add a verification result and rollback action to the migration step.
  • Keep the existing service name, command text, and evidence references unchanged.

「stow」を使用しています。 Check a handoff artifact against its schema and report the result.

期待される結果:

The report separates parse failures, schema violations, unresolved references, and advisory prose concerns so the owner can correct the artifact.

セキュリティ監査

安全
v1 • 2026/9/6 バージョン別レポートを開く

All 254 static matches are false positives in their reviewed context. They are documentation literals, schema identifiers, parser patterns, or local validation references; no command execution, network request, secret access, destructive file operation, or encoded payload was evidenced. No prompt-injection or data-exfiltration intent was found.

78
スキャンされたファイル
13,928
解析済み行数
0
レビュー項目
0
誤検知を無視

リスク要因

⚙️ 外部コマンド (50)
corpus/action-shaping.md:24-25 corpus/action-shaping.md:25-26 corpus/action-shaping.md:26-27 rules/conflicts.yaml:7 rules/conflicts.yaml:11 rules/conflicts.yaml:96 rules/conflicts.yaml:97 rules/conflicts.yaml:249 runtime/dictionary_lookup.py:532 runtime/lint_prose.py:8 runtime/lint_prose.py:14 runtime/lint_prose.py:15-16 runtime/lint_prose.py:16 runtime/lint_prose.py:23 runtime/lint_prose.py:25 runtime/lint_prose.py:30 runtime/lint_prose.py:31 runtime/lint_prose.py:35 runtime/lint_prose.py:70 runtime/lint_prose.py:71 runtime/lint_prose.py:72 runtime/lint_prose.py:105 runtime/lint_prose.py:122 runtime/lint_prose.py:141 runtime/lint_prose.py:166 runtime/lint_prose.py:175 runtime/lint_prose.py:177 runtime/lint_prose.py:185 runtime/lint_prose.py:248 runtime/lint_prose.py:276 runtime/lint_prose.py:278 runtime/lint_prose.py:353 runtime/lint_prose.py:529 runtime/lint_prose.py:628 runtime/lint_prose.py:700 runtime/lint_prose.py:702 runtime/lint_prose.py:704 runtime/lint_prose.py:705 runtime/profiles.py:9 runtime/profiles.py:13 runtime/profiles.py:14 runtime/profiles.py:62 runtime/profiles.py:63-64 runtime/profiles.py:64-82 runtime/profiles.py:82 runtime/profiles.py:84 runtime/query_rules.py:48 runtime/query_rules.py:125 runtime/query_rules.py:127 runtime/query_rules.py:128
🔑 環境変数 (2)
rules/conflicts.yaml:191 rules/conflicts.yaml:193
🌐 ネットワークアクセス (21)
rules/controlled-dictionary.schema.json:2 rules/project-terminology.schema.json:2 rules/registry.schema.json:2 rules/registry.schema.json:3 runtime/lint_prose.py:106 schemas/event.schema.json:2 schemas/event.schema.json:3 schemas/evidence-record.schema.json:2 schemas/evidence-record.schema.json:3 schemas/handoff.schema.json:2 schemas/handoff.schema.json:3 schemas/output-contract.schema.json:2 schemas/output-contract.schema.json:3 schemas/plan.schema.json:2 schemas/plan.schema.json:3 schemas/runbook.schema.json:2 schemas/runbook.schema.json:3 schemas/state.schema.json:2 schemas/state.schema.json:3 schemas/task-packet.schema.json:2 schemas/task-packet.schema.json:3
📁 ファイルシステムへのアクセス (6)
rules/registry.yaml:2405 runtime/lint_prose.py:107 runtime/query_rules.py:199 runtime/query_rules.py:27 runtime/validate.py:541 runtime/validate.py:328
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: claude
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/theislampill-stow/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/theislampill-stow/security.svg)](https://skillstore.io/skills/theislampill-stow?utm_source=security_passport_badge)

HTMLバッジ

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

埋め込みカード

<iframe src="https://skillstore.io/embed/skills/theislampill-stow.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

theislampill. (2026). stow security audit report (audit version 1) [Author version 0.4.2]. Skillstore. https://skillstore.io/skills/theislampill-stow/audits/1

BibTeX形式の引用

@techreport{theislampill-theislampill-stow-2026, author = {theislampill}, title = {stow security audit report (audit version 1)}, institution = {Skillstore}, year = {2026}, number = {1}, url = {https://skillstore.io/skills/theislampill-stow/audits/1}, note = {Author version 0.4.2} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "stow security audit report (audit version 1)" version: "0.4.2" type: report authors: - name: "theislampill" date-released: "2026-09-06" url: "https://skillstore.io/skills/theislampill-stow/audits/1" identifiers: - type: other value: "skillstore:theislampill-stow: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
45
アーキテクチャ
90
保守性
87
コンテンツ
65
コミュニティ
100
仕様準拠

作成できるもの

Polish technical documentation

Rewrite a README or public technical document for direct structure, precise claims, consistent terminology, and preserved command examples.

Prepare operational procedures

Shape runbooks, plans, audits, and handoffs into bounded steps with visible evidence, limitations, verification, and rollback details.

Validate structured deliverables

Check JSON, JSONL, YAML, or Markdown artifacts against their contracts while retaining literal values and reporting structural defects.

これらのプロンプトを試す

Clean up a README 
Rewrite this README for direct technical clarity. Preserve every command, path, identifier, quotation, and data value. Return only the revised README.
Review a runbook 
Review this runbook for missing preconditions, unclear actions, unsupported claims, and absent verification or rollback details. List concrete corrections and preserve all commands exactly.
Check a structured artifact 
Validate this JSON, JSONL, YAML, or Markdown artifact against the stated contract. Report parse and structural issues, distinguish warnings from failures, and do not rewrite protected values.
Audit rule applicability 
Apply STOW to this technical document. Classify each region, identify the active profile and relevant rule conflicts, preserve protected literals, and return the smallest complete set of corrections with evidence.

ベストプラクティス

  • State the output contract and protected literals before requesting a rewrite.
  • Provide authoritative sources for claims, numbers, terminology, and required formats.
  • Use the named validator on the actual final artifact when the format has a supported schema.

回避

  • Do not ask STOW to invent commands, evidence, versions, citations, or current system state.
  • Do not treat advisory prose findings as proof of a security, correctness, or authorship defect.
  • Do not label editable text as protected when the requested change requires literal editing.

よくある質問

What is STOW?
STOW is a contract-first assistant for precise technical writing and validated structured artifacts.
Which tools support STOW?
The skill supports Claude, Codex, and Claude Code compatible hosts.
What content can STOW review?
It reviews READMEs, runbooks, procedures, plans, audits, handoffs, mixed prose, and JSON, JSONL, YAML, or Markdown artifacts.
Does STOW change commands or paths?
It preserves protected commands, paths, identifiers, quotations, and data values unless literal editing is explicitly requested.
Are prose lint results blocking?
The packaged prose linter is report-only. Its advisories support review and do not replace human judgment.
Can STOW validate schemas?
Yes. Its packaged validator checks supported structured formats and named coordination schemas when the required dependencies are installed.

開発者情報

作成者

theislampill

ライセンス

LICENSE

作者バージョン

v0.4.2

Skillstore リビジョン

r1

リポジトリ

https://github.com/theislampill/stow/tree/bab13dbd8bd83cfc8cc72b7f64761e668302ff6e/skills/stow

参照

e5464e662405de6361fdde6b984f9ea865635f64

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

📁 corpus/

📄 action-shaping.md

📄 descriptions.md

📄 general-practice.md

📄 multiword-nouns.md

📄 procedures.md

📁 prose-integrity/

📄 banned-lists.md

📄 detection-patterns.md

📄 examples.md

📄 false-positives.md

📄 rules.md

📄 punctuation.md

📄 safety.md

📄 sentences.md

📁 style/

📄 consistency.md

📄 economy.md

📁 verbs/

📄 technical-verbs.md

📄 verb-forms.md

📁 words/

📄 approved-tables.md

📄 selection.md

📄 usage.md

📄 LICENSE

📁 references/

📄 action-shaping.md

📄 activation-and-precedence.md

📄 agent-handoffs.md

📄 ai-writing-detection.md

📄 always-on.md

📄 audit-and-evidence.md

📄 canonical-terms.md

📄 conformance.md

📄 continuity-and-state.md

📄 controlled-technical-writing.md

📄 cross-harness-interchange.md

📄 descriptions.md

📄 descriptive-prose.md

📄 format-json.md

📄 format-jsonl.md

📄 format-markdown.md

📄 format-yaml.md

📄 implementation-plans.md

📄 meta-code.md

📄 procedures.md

📄 prose-integrity.md

📄 protected-regions.md

📄 public-documentation.md

📄 rule-index.md

📄 runbooks.md

📄 safety-instructions.md

📄 technical-clarity.md

📁 rules/

📄 artifact-classes.yaml

📄 conflicts.yaml

📄 controlled-dictionary-v1.json.gz

📄 controlled-dictionary.schema.json

📄 profiles.json

📄 project-terminology.schema.json

📄 registry.schema.json

📄 registry.yaml

📄 routing.yaml

📁 runtime/

📄 dictionary_lookup.py

📄 lint_prose.py

📄 profiles.py

📄 query_rules.py

📄 validate_terms.py

📄 validate.py

📁 schemas/

📄 event.schema.json

📄 evidence-record.schema.json

📄 handoff.schema.json

📄 output-contract.schema.json

📄 plan.schema.json

📄 runbook.schema.json

📄 state.schema.json

📄 task-packet.schema.json

📄 SKILL.md

📁 templates/

📄 AUDIT.md

📄 event-stream.jsonl

📄 HANDOFF.md

📄 PLAN.md

📄 RUNBOOK.md

📄 STATE.md

📄 task-packet.yaml