技能 chatkit-botbuilder

📦

chatkit-botbuilder

Name: chatkit-botbuilder
Author: 92Bilal26

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

建置生產環境 ChatKit 聊天機器人

團隊需要可靠的模式，將 ChatKit 連接到真實應用程式工作流程。此技能提供 FastAPI 後端、MCP 工具、Next.js 小工具與使用者隔離的架構指引。

支援: Claude Codex Code(CC)

⚠️ 38 差

Maintain this Skill? Claim it ⚠️ 38 差

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 "chatkit-botbuilder" from https://skillstore.io/skills/92bilal26-chatkit-botbuilder.md and its manifest at https://skillstore.io/api/skills/92bilal26-chatkit-botbuilder/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/92bilal26-chatkit-botbuilder.md 已簽名 manifest GET /api/skills/92bilal26-chatkit-botbuilder/manifest 已簽名 lockfile GET /api/skills/92bilal26-chatkit-botbuilder/lockfile

測試它

正在使用「chatkit-botbuilder」。我需要一個用於管理客戶支援工單的 ChatKit 聊天機器人。

預期結果:

採用分層架構，包含 Next.js ChatKit 面板、FastAPI protocol endpoint，以及工單專用 MCP wrappers。
一份檢查清單，涵蓋 JWT 擷取、工單擁有權 filters、串流回應與儀表板重新整理行為。

正在使用「chatkit-botbuilder」。我的 ChatKit tool 為錯誤的使用者建立資料。

預期結果:

一條診斷路徑，用於檢查 middleware 使用者擷取、wrapper closure scope 與資料庫 filters。
一個修正後的實作計畫，將 user_id 保留在伺服器端，並套用到每一次讀取與寫入。

正在使用「chatkit-botbuilder」。準備將此 ChatKit 整合部署上線。

預期結果:

一份部署檢查清單，涵蓋 domain registration、backend URL configuration、CORS、secret placement 與 production token validation。
一份發布前需審查的安全項目清單，包含 frontend secrets 與詳細錯誤回應。

安全審計

高風險

v6 • 2026/7/5 開啟版本化報告

Most static findings are false positives caused by Markdown code examples, inline backticks, localhost URLs, and ordinary architecture terms. The confirmed security issue is guidance that places an OpenAI API key in frontend public configuration, with additional semantic concerns around browser token storage and development test-token bypasses. No prompt injection attempt was found in the reviewed skill files.

已掃描檔案

2,579

分析行數

複核項目

已忽略的誤報

已確認的安全問題 (3)

高

CHATKIT_SKILL_SUMMARY.md:216-222 references/frontend_integration.md:385-398

Frontend OpenAI API Key Exposure Guidance

The documentation recommends NEXT_PUBLIC_OPENAI_API_KEY in frontend configuration. In Next.js, NEXT_PUBLIC values are exposed to the browser, so copying this guidance can leak an OpenAI API key.

The exact variable name appears in frontend environment guidance and includes an sk-style placeholder in the detailed integration guide. The risk follows directly from Next.js public environment variable behavior.

中

references/frontend_integration.md:31-45 references/frontend_integration.md:326-327

Browser Storage for Bearer Tokens

The frontend examples read and write JWT bearer tokens in localStorage and sessionStorage. This pattern can expose sessions to XSS and should be replaced or mitigated for production deployments.

The cited snippets explicitly store and retrieve access tokens from browser storage. This is a known session handling risk, although some applications may accept it with strong XSS controls.

中

references/backend_architecture.md:34-38

Development Test Token Bypass Pattern

The backend example accepts bearer tokens that start with test-token when the environment is development. This is acceptable for local testing but can become an authentication bypass if the environment gate is misconfigured.

The bypass is explicitly scoped to development, so malicious intent is not evident. The pattern is still risky guidance because a copied or misconfigured deployment could accept forged user IDs.

能力複核項目 (2)

這些是真實的本機能力，對此技能而言可能是預期行為，因此需要複核，但不會按已確認的惡意行為計分。

高

CHATKIT_SKILL_SUMMARY.md:219

Generic API/secret keys

- `NEXT_PUBLIC_OPENAI_API_KEY` = Optional (if frontend needs direct OpenAI access)

The guidance suggests NEXT_PUBLIC_OPENAI_API_KEY for frontend use, which would expose an OpenAI secret in browser-visible code if copied. This is a real credential exposure risk, not just a placeholder mention.

高

references/frontend_integration.md:397

Generic API/secret keys

NEXT_PUBLIC_OPENAI_API_KEY=sk-your-key-here

The .env.local example places NEXT_PUBLIC_OPENAI_API_KEY with an sk-style placeholder in frontend configuration. NEXT_PUBLIC variables are bundled for the browser, so this pattern can leak API credentials.

風險因素

🌐 網路存取 (15)

CHATKIT_SKILL_SUMMARY.md:186 CHATKIT_SKILL_SUMMARY.md:187 references/backend_architecture.md:333 references/frontend_integration.md:72 references/frontend_integration.md:271 references/frontend_integration.md:19 references/frontend_integration.md:183 references/frontend_integration.md:391 references/frontend_integration.md:404 references/frontend_integration.md:410 references/user_isolation.md:142 references/user_isolation.md:146 references/user_isolation.md:151 references/user_isolation.md:447 SKILL.md:172

📁 檔案系統存取 (2)

CHATKIT_SKILL_SUMMARY.md:7 CHATKIT_SKILL_SUMMARY.md:7

🔑 環境變數 (9)

CHATKIT_SKILL_SUMMARY.md:201 CHATKIT_SKILL_SUMMARY.md:201 CHATKIT_SKILL_SUMMARY.md:219 CHATKIT_SKILL_SUMMARY.md:222 references/frontend_integration.md:19 references/frontend_integration.md:20 references/frontend_integration.md:19 references/frontend_integration.md:20 references/frontend_integration.md:397

⚙️ 外部命令 (51)

references/frontend_integration.md:23-66 references/frontend_integration.md:66-176 references/frontend_integration.md:273-285 references/frontend_integration.md:285-313 references/user_isolation.md:85 references/user_isolation.md:254 references/user_isolation.md:264 SKILL.md:41-59 SKILL.md:59-98 SKILL.md:98-112 SKILL.md:112-116 SKILL.md:116-123 SKILL.md:123-127 SKILL.md:127-134 SKILL.md:134-138 SKILL.md:138-144 SKILL.md:144-150 SKILL.md:150-153 SKILL.md:153-161 SKILL.md:161-165 SKILL.md:165-170 SKILL.md:170-174 SKILL.md:174-178 SKILL.md:178-191 SKILL.md:191-195 SKILL.md:195-209 SKILL.md:209-215 SKILL.md:215-229 SKILL.md:229-233 SKILL.md:233-239 SKILL.md:239-253 SKILL.md:253-263 SKILL.md:263-267 SKILL.md:267-300 SKILL.md:300-304 SKILL.md:304-315 SKILL.md:315-319 SKILL.md:319-334 SKILL.md:334-384 SKILL.md:384-387 SKILL.md:387-394 SKILL.md:394-399 SKILL.md:399-406 SKILL.md:406-409 SKILL.md:409-416 SKILL.md:416-418 SKILL.md:418-419 SKILL.md:419-435 SKILL.md:435-446 SKILL.md:446-507 SKILL.md:507

⚡ 包含腳本 (4)

references/mcp_wrapper_guide.md:183-190 references/user_isolation.md:85 references/user_isolation.md:254 references/user_isolation.md:264

審計單位: codex 檢視審計紀錄 →

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/92bilal26-chatkit-botbuilder/audits/6?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/92bilal26-chatkit-botbuilder/security.svg)](https://skillstore.io/skills/92bilal26-chatkit-botbuilder?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/92bilal26-chatkit-botbuilder?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/92bilal26-chatkit-botbuilder/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/92bilal26-chatkit-botbuilder.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

92Bilal26. (2026). chatkit-botbuilder security audit report (audit version 6) [Author version unspecified]. Skillstore. https://skillstore.io/skills/92bilal26-chatkit-botbuilder/audits/6

BibTeX 引用

@techreport{92bilal26-92bilal26-chatkit-botbuilder-2026,
  author = {92Bilal26},
  title = {chatkit-botbuilder security audit report (audit version 6)},
  institution = {Skillstore},
  year = {2026},
  number = {6},
  url = {https://skillstore.io/skills/92bilal26-chatkit-botbuilder/audits/6},
  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: "chatkit-botbuilder security audit report (audit version 6)"
version: "unspecified"
type: report
authors:
  - name: "92Bilal26"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/92bilal26-chatkit-botbuilder/audits/6"
identifiers:
  - type: other
    value: "skillstore:92bilal26-chatkit-botbuilder:audit:6"
    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

建立內部任務助理

設計一個 ChatKit 助理，用於建立、列出、更新並完成使用者專屬任務。

將 ChatKit 加入現有儀表板

規劃儀表板整合所需的前端小工具、已驗證 API 呼叫與後端 endpoint。

審查多使用者隔離

檢查 middleware、wrappers 與資料庫 filters 是否維持每位使用者的資料邊界。

試試這些提示

規劃基本 ChatKit 應用程式

Use this skill to outline a simple ChatKit chatbot for a task app. Include the frontend, backend, and tool layers.

與我的後端整合

Use this skill to adapt my FastAPI app for ChatKit. Focus on the protocol endpoint, streaming response, and JWT context.

設計 MCP 工具封裝

Use this skill to design MCP wrapper functions for my app tools. Ensure every tool call receives the authenticated user_id.

稽核生產就緒度

Use this skill to review my ChatKit architecture for production. Check user isolation, token handling, error handling, CORS, and deployment settings.

最佳實務

將 secrets 保留在後端，只向瀏覽器公開 public configuration values。
從已驗證的伺服器端 authentication context 注入 user_id，而不是從 request bodies 或 model output 注入。
使用至少兩個不同使用者與衝突的 resource IDs 進行測試，以驗證使用者隔離。