chatgpt-app-sdk
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate documentation, code patterns, and architectural guidance for developers building on the OpenAI Apps SDK platform. All provided code snippets, such as the React hooks for state management and the Vite configuration for UI components, follow official patterns and security best practices.
- [SAFE]: External references and links are restricted to official and well-known documentation sites, including modelcontextprotocol.io, openai.github.io, and the official Model Context Protocol GitHub repositories.
- [SAFE]: The documentation includes proactive security advice, such as enforcing server-side authentication, configuring Content Security Policy (CSP), and avoiding the inclusion of secrets in model-visible payloads.
- [SAFE]: Recommendations for local development tools, such as using Cloudflare Tunnels (cloudflared), are standard industry practices and are documented with appropriate security measures (e.g., file permission hardening).
Audit Metadata