chatgpt-app-builder

Malicious tool definition detected

### Claude Guide user to submit the app on the [Anthropic Connectors Directory FAQ](https://support.claude.com/en/articles/11596036-anthropic-connectors-directory-faq).

Malicious tool definition detected

**Configure**: - **Branch**: pick the branch to deploy from (pushes trigger redeploys) - **Environment variables** - **Build settings**: Alpic auto-detects, verify build command and output directory 5.

Malicious tool definition detected

Discovery Endpoints Serve two JSON endpoints so the LLM host knows where to authenticate users: ```typescript // index.ts import { mcpAuthMetadataRouter } from "@modelcontextprotocol/sdk/server/auth/router.js"; app.use( mcpAuthMetadataRouter({ oauthMetadata: { issuer: "https://your-oauth-provider.com", authorization_endpoint: "https://your-oauth-provider.com/oauth2/authorize", token_endpoint: "https://your-oauth-provider.com/oauth2/token", response_types_supported: ["code"], grant_types_supporte

Malicious tool definition detected

If yes, expose the local server via ngrok: ```bash ngrok http 3000 ``` Extract the forwarding URL from ngrok output (e.g., `https://abc123.ngrok.io`).

Tool passed security scan

Tool passed security scan

Tool passed security scan

Tool passed security scan