技能 gtm-workflow

📦

gtm-workflow

Name: gtm-workflow
Author: eliasstravik

內容修訂 r1 中風險 🔑 環境變數🌐 網路存取📁 檔案系統存取⚡ 包含腳本⚙️ 外部命令

Build and manage GTM workflows

Reusable GTM work needs consistent setup, testing, approvals, and run tracking. This skill creates and maintains workflow projects with cost-aware checks and deployment steps.

支援: Claude Codex Code(CC)

📊 72 尚可

Maintain this Skill? Claim it 📊 72 尚可

安裝前請停止並請求確認。

請審查安裝計畫，並在變更檔案前取得使用者明確同意。

使用我的智慧代理安裝

將此請求複製給你的智慧代理，其中包含權威 Skill 頁面和清單。

智能體請求

Review the Skillstore skill "gtm-workflow" from https://skillstore.io/skills/eliasstravik-gtm-workflow.md and its manifest at https://skillstore.io/api/skills/eliasstravik-gtm-workflow/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

你的智慧代理仍應顯示其計畫，並依安全政策請求必要確認。

Agent 可讀資源

當 AI Agent、爬蟲或腳本需要乾淨脈絡、而不是讀取完整頁面時，請使用這些連結。

Markdown 詳情 GET /skills/eliasstravik-gtm-workflow.md 已簽名 manifest GET /api/skills/eliasstravik-gtm-workflow/manifest 已簽名 lockfile GET /api/skills/eliasstravik-gtm-workflow/lockfile

測試它

正在使用「gtm-workflow」。 Create a workflow for a small lead list.

預期結果:

Workflow plan with inputs, result table, and one-row test
Approval card describing saved changes, cost, and external effects
Verification result and workflow diagram after testing

正在使用「gtm-workflow」。 Check the latest deployed workflow run.

預期結果:

Run status and completion summary
Recorded cost and row outcome counts
Plain-language failure reason when a run does not complete

安全審計

中風險

v1 • 2026/9/11 開啟版本化報告

The 400 static matches are predominantly scanner false positives: lockfile integrity hashes, ordinary key identifiers, documentation, and bounded workflow tooling. No hidden payload, prompt injection, credential exfiltration, or command injection was found in the reviewed context. The optional installer persistently modifies user-level Claude Code hook configuration and should require explicit informed consent. Static review was capped at 400/1755 representative findings; omitted static matches are unconfirmed, so automatic publishing stays disabled until manual review.

已掃描檔案

15,404

分析行數

複核項目

已忽略的誤報

已確認的安全問題 (1)

中

SKILL.md:53 scripts/command-permission.mjs:162-172

Persistent global hook installation

The optional installer writes PreToolUse hooks into the user-level Claude Code settings file. This creates persistent command interception outside an individual workflow project and warrants explicit informed consent.

The skill explicitly instructs users to run the installer, and the installer writes hook entries to ~/.claude/settings.json. The behavior is transparent and opt-in, but it is persistent global configuration.

風險因素

🔑 環境變數 (50)

🌐 網路存取 (50)

📁 檔案系統存取 (50)

references/library.md:23 references/library.md:24 references/library.md:25 references/library.md:26 references/library.md:27 scripts/command-permission.mjs:4 scripts/command-permission.mjs:4 scripts/command-permission.mjs:24 scripts/command-permission.mjs:33 scripts/command-permission.mjs:82 scripts/command-permission.mjs:125 scripts/command-permission.mjs:141 scripts/command-permission.mjs:133 scripts/setup-workflow-project.sh:19 scripts/setup-workflow-project.sh:4 scripts/setup-workflow-project.sh:10 scripts/setup-workflow-project.sh:13 SKILL.md:38 SKILL.md:69 templates/.gitignore:1 templates/lib/agent-stage.ts:1 templates/lib/agent-stage.ts:73 templates/lib/agent.ts:103 templates/lib/agent.ts:2 templates/lib/agent.ts:85 templates/lib/diagram-svg.ts:2 templates/lib/diagram-svg.ts:112 templates/lib/diagram-svg.ts:122 templates/lib/diagram-svg.ts:112 templates/lib/diagram-svg.ts:2 templates/lib/diagram-svg.ts:119 templates/lib/rows.ts:239 templates/package-lock.json:2528 templates/package-lock.json:2531 templates/package-lock.json:3430 templates/package-lock.json:3450 templates/package-lock.json:3452 templates/package-lock.json:3613 templates/package-lock.json:3626 templates/package-lock.json:3628 templates/package-lock.json:4885 templates/package-lock.json:5020 templates/package-lock.json:5022 templates/package-lock.json:9408 templates/package-lock.json:9463 templates/package-lock.json:4106 templates/package-lock.json:9030 templates/package-lock.json:9032 templates/package.json:8 templates/scripts/generate-migration.mjs:10

⚡ 包含腳本 (12)

scripts/command-permission.mjs:92 scripts/setup-workflow-project.sh:20 scripts/setup-workflow-project.sh:21 templates/lib/db.ts:14 templates/lib/db.ts:14 templates/lib/db.ts:14 templates/lib/db.ts:23 templates/lib/db.ts:58 templates/lib/db.ts:149 templates/package.json:12 templates/scripts/gtm.ts:48 templates/scripts/gtm.ts:72

⚙️ 外部命令 (50)

審計單位: claude

分享與引用此報告

分享版本化評估報告、中性徽章、嵌入卡片與引用。Skillstore 只報告證據，不判斷該 Skill 是否安全。

開啟版本化報告

複製報告連結

https://skillstore.io/skills/eliasstravik-gtm-workflow/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

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

HTML 徽章

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

嵌入卡片

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

學術引用 (APA · BibTeX · CFF)

APA 引用

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

BibTeX 引用

@techreport{eliasstravik-eliasstravik-gtm-workflow-2026,
  author = {eliasstravik},
  title = {gtm-workflow security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/eliasstravik-gtm-workflow/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: "gtm-workflow security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "eliasstravik"
date-released: "2026-09-11"
url: "https://skillstore.io/skills/eliasstravik-gtm-workflow/audits/1"
identifiers:
  - type: other
    value: "skillstore:eliasstravik-gtm-workflow:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 評分

評分依據證據置信度: 中

架構

可維護性

內容

社群

規範符合度

你可以打造什麼

Create a lead enrichment workflow

Build a repeatable workflow that validates rows, enriches records, and stores results after a one-row test.

Operate a deployed workflow

Inspect recent runs, query stored results, approve a checkpoint, or cancel an active workflow.

Maintain workflow infrastructure

Add a migration, update a schedule, verify deployment readiness, and prepare a controlled release.

試試這些提示

Create a first workflow

Create a reusable workflow that validates a small list of leads. Include a one-row test and explain the approval needed before running it.

Inspect a run

Inspect the latest run for the lead-enrichment workflow. Summarize its status, cost, completed rows, and failures.

Add a safe schedule

Add a weekday schedule to the account-refresh workflow. Verify the workflow and show the external effects before deployment.

Update with controls

Update the enrichment workflow to add an agentic research stage. Set a per-row spend cap, add Diagram cost metadata, test one row, and request approval before deployment.

最佳實務

Use a one-row test and review its projected cost before every new workflow run.
Keep credentials in environment variables and do not commit local data or environment files.
Review migration effects and request fresh approval when costs or external effects change.

避免

Do not use this skill for one-off work that should not become a reusable workflow.
Do not run a full workflow before verification and explicit approval.
Do not store secrets, test data, or generated runtime files in version control.