firebase-launchpad

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses the local Expo CLI state file (~/.expo/state.json) to facilitate authentication with Expo's GraphQL API. This is standard behavior for automation tools targeting the Expo ecosystem and is used for legitimate authentication purposes.
  • [PROMPT_INJECTION]: The skill reads project configuration from app.json to determine parameters for API requests, which represents a surface for indirect prompt injection. 1. Ingestion points: app.json (project metadata). 2. Boundary markers: None identified in the provided templates. 3. Capability inventory: Shell execution via pnpm, file system writing, and network operations to Google and Expo APIs. 4. Sanitization: Not explicitly implemented; relies on valid project data.
  • [COMMAND_EXECUTION]: The skill executes locally generated or template-based JavaScript files using node and pnpm to perform its provisioning tasks.
  • [EXTERNAL_DOWNLOADS]: Downloads the googleapis library from the public npm registry and uses resources from the author's own repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 11:34 PM