技能 chatgpt-app-builder
📦

chatgpt-app-builder

内容修订 r1 安全 📁 文件系统访问⚙️ 外部命令🌐 网络访问🔑 环境变量

Build ChatGPT Apps with Skybridge

Building a ChatGPT app requires coordinated MCP tools, React views, authentication, testing, and deployment. This skill guides the complete Skybridge workflow from discovery through publishing.

支持: Claude Codex Code(CC)
📊 75 合格

使用我的智能体安装

将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。

智能体请求
Review the Skillstore skill "chatgpt-app-builder" from https://skillstore.io/skills/alpic-ai-chatgpt-app-builder.md and its manifest at https://skillstore.io/api/skills/alpic-ai-chatgpt-app-builder/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智能体仍应展示其计划，并按安全策略请求必要确认。

Agent 可读资源

当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。

Markdown 详情 GET /skills/alpic-ai-chatgpt-app-builder.md 已签名 manifest GET /api/skills/alpic-ai-chatgpt-app-builder/manifest 已签名 lockfile GET /api/skills/alpic-ai-chatgpt-app-builder/lockfile

测试它

正在使用“chatgpt-app-builder”。 Design a flight search app from an approved specification.

预期结果:

  • A search tool accepts destination and dates.
  • A results view compares flights and preserves selection state.
  • A checkout tool returns an approved external redirect.

正在使用“chatgpt-app-builder”。 Add authentication while keeping public search available.

预期结果:

The app uses a configured OAuth provider, leaves search anonymous, protects account tools, and checks required scopes before returning user data.

正在使用“chatgpt-app-builder”。 Prepare the app for directory submission.

预期结果:

The review verifies annotations, declared CSP domains, ownership setup, responsive display modes, authentication behavior, and successful local evaluation results.

安全审计

安全
v1 • 2026/9/14 打开版本化报告

All 89 static findings are false positives caused by Markdown syntax, TypeScript template literals, relative imports, placeholder URLs, local development endpoints, or standard OAuth configuration. No prompt injection, credential exfiltration, unsafe path handling, command execution, or malicious intent was found.

28
已扫描文件
2,306
分析行数
0
复核项
0
已忽略误报

风险因素

📁 文件系统访问 (8)
evals/skill.json:28 references/ecommerce.md:232 references/evals.md:23 references/fetch-and-render-data.md:124 references/oauth.md:8 references/oauth.md:147 references/oauth.md:235 references/publish.md:16
⚙️ 外部命令 (27)
references/architecture.md:159-160 references/architecture.md:160-162 references/architecture.md:162-165 references/architecture.md:165-166 references/architecture.md:166-169 references/architecture.md:169-170 references/architecture.md:170-173 references/architecture.md:173-174 references/architecture.md:174-175 references/download-file.md:18-38 references/fetch-and-render-data.md:12-19 references/fetch-and-render-data.md:71-87 references/fetch-and-render-data.md:87-94 references/fetch-and-render-data.md:129-163 references/oauth.md:25-31 references/oauth.md:97-101 references/oauth.md:206-216 references/oauth.md:234-235 references/oauth.md:235-244 references/open-external-links.md:19-24 references/state-and-context.md:76-82 references/state-and-context.md:101-102 references/state-and-context.md:135-136 references/state-and-context.md:136-154 references/ui-guidelines.md:134-139 SKILL.md:23 SKILL.md:25
🌐 网络访问 (21)
references/csp.md:22 references/csp.md:23 references/csp.md:24 references/csp.md:25 references/download-file.md:64 references/ecommerce.md:156 references/ecommerce.md:211 references/migrate-to-v1.md:113 references/oauth.md:76 references/oauth.md:147 references/oauth.md:152 references/oauth.md:184 references/oauth.md:185 references/oauth.md:186 references/open-external-links.md:19 references/open-external-links.md:36 references/open-external-links.md:47 references/open-external-links.md:65 SKILL.md:25 SKILL.md:52 SKILL.md:54
🔑 环境变量 (5)
references/evals.md:12 references/oauth.md:190 references/oauth.md:235 references/oauth.md:190 references/oauth.md:235
最新一次完成的静态与语义审计未检测到已确认的安全发现。这并不能证明该 Skill 没有副作用。
审计者: codex
分享与引用此报告

分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。

打开版本化报告
安全评估

复制报告链接

https://skillstore.io/skills/alpic-ai-chatgpt-app-builder/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/alpic-ai-chatgpt-app-builder/security.svg)](https://skillstore.io/skills/alpic-ai-chatgpt-app-builder?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/alpic-ai-chatgpt-app-builder?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/alpic-ai-chatgpt-app-builder/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/alpic-ai-chatgpt-app-builder.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
学术引用 (APA · BibTeX · CFF)

APA 引用

alpic-ai. (2026). chatgpt-app-builder security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/alpic-ai-chatgpt-app-builder/audits/1

BibTeX 引用

@techreport{alpic-ai-alpic-ai-chatgpt-app-builder-2026, author = {alpic-ai}, title = {chatgpt-app-builder security audit report (audit version 1)}, institution = {Skillstore}, year = {2026}, number = {1}, url = {https://skillstore.io/skills/alpic-ai-chatgpt-app-builder/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: "chatgpt-app-builder security audit report (audit version 1)" version: "unspecified" type: report authors: - name: "alpic-ai" date-released: "2026-09-14" url: "https://skillstore.io/skills/alpic-ai-chatgpt-app-builder/audits/1" identifiers: - type: other value: "skillstore:alpic-ai-chatgpt-app-builder:audit:1" description: "Skillstore immutable audit report identifier"

比较变体

4 个可安装变体

每位作者的技能都保持为独立的可安装项。推荐变体按 Skillstore 证据排序。

为什么此变体排名第一

Skillstore 评分最高
mcp-use 推荐

mcp-use-chatgpt-app-builder

Skillstore 评分 76
证据置信度 高
Skillstore 使用量 11
已更新

2026-09-09

alpic-ai 当前

alpic-ai-chatgpt-app-builder

Skillstore 评分 75
证据置信度 中
Skillstore 使用量 2
已更新

2026-09-22

shubhamsaboo

shubhamsaboo-chatgpt-app-builder

Skillstore 评分 67
证据置信度 中
Skillstore 使用量 6
已更新

2026-09-09

BayramAnnakov

bayramannakov-chatgpt-app-builder

Skillstore 评分 38
证据置信度 中
Skillstore 使用量 12
已更新

2026-09-09

Skillstore 评分

评分依据 证据置信度: 中
45
架构
85
可维护性
87
内容
65
社区
83
规范符合性

你能构建什么

Prototype a ChatGPT app

Turn a focused product idea into a specification, MCP tool design, and interactive React view.

Add secure user data access

Configure OAuth, tool permissions, scopes, and authenticated handlers for user-specific workflows.

Prepare an app for release

Run local checks, model evaluations, deployment steps, CSP audits, and directory submission tasks.

试试这些提示

Explore an app idea 
Help me define a ChatGPT app for [audience]. The main problem is [problem]. Ask one discovery question at a time.
Design tools and views 
Review my SPEC.md. Design focused MCP tools and React views for [workflow]. Explain which steps need custom UI.
Implement authenticated data access 
Add [OAuth provider] authentication to this Skybridge app. Protect [tools] with [scopes] and keep [public tool] anonymous.
Validate and release the app 
Audit this Skybridge app for tool annotations, CSP, responsive views, authentication, evaluations, deployment, and directory submission. Fix verified issues.

最佳实践

  • Keep SPEC.md current before changing architecture or implementation.
  • Expose concise model data while keeping large display-only payloads in view metadata.
  • Verify authentication, tool annotations, CSP domains, and responsive layouts before deployment.

避免

  • Do not port an entire application when a few focused conversational actions are sufficient.
  • Do not place secrets in model-visible content, view metadata, logs, or committed files.
  • Do not duplicate the same payload across content, structured data, view state, and model annotations.

常见问题

What does this skill build?
It guides development of ChatGPT apps and MCP servers with Skybridge, TypeScript, and React views.
Does it support existing projects?
Yes. It can update a project with an existing SPEC.md and guide supported Skybridge migrations.
Can it add authentication?
Yes. It covers built-in OAuth providers, custom providers, mixed public and protected tools, scopes, and manual verification.
How are applications tested?
The workflow supports local DevTools checks and behavior-focused evaluations that call a real language model.
Which deployment path is documented?
The skill documents Alpic CLI deployment, local tunneling, and connection steps for ChatGPT and Claude.
Will evaluations use paid services?
They can. Live evaluations require a supported model provider, an API key, and may create usage charges.

开发者详情

作者

alpic-ai

许可证

MIT

Skillstore 修订

r1

版本提示

作者未声明版本。

仓库

https://github.com/alpic-ai/skybridge/tree/8fd1fe87aa28c89375afd21cf42b05e6bddd0f52/skills/chatgpt-app-builder

引用

06161a591a23cfa999283946c003cdd91f2317df

维护新鲜度

2026/9/14

使用量

0 次下载 · 0 次浏览

文件结构

📁 evals/

📄 architecture.json

📄 communicate-with-host.json

📄 discover.json

📄 fetch-and-render-data.json

📄 open-external-links.json

📄 README.md

📄 skill.json

📄 state-and-context.json

📄 ui-guidelines.json

📄 update-existing.json

📁 references/

📄 architecture.md

📄 copy-template.md

📄 csp.md

📄 deploy.md

📄 discover.md

📄 download-file.md

📄 ecommerce.md

📄 evals.md

📄 fetch-and-render-data.md

📄 migrate-to-v1.md

📄 oauth.md

📄 open-external-links.md

📄 prompt-llm.md

📄 publish.md

📄 run-locally.md

📄 state-and-context.md

📄 ui-guidelines.md

📄 SKILL.md