google-contacts

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes contact records retrieved from the Google People API, which creates a surface for indirect prompt injection if records contain malicious instructions. This risk is inherent to the contact management use-case and is managed by structural validation.\n
  • Ingestion points: Contacts are fetched via tools in listContacts.ts, searchContacts.ts, getContact.ts, listOtherContacts.ts, and searchOtherContacts.ts.\n
  • Boundary markers: The skill instructions do not currently include explicit natural language markers to warn the agent against executing instructions found within contact data.\n
  • Capability inventory: The skill's capabilities are restricted to network requests to Google API endpoints via ctx.fetch.\n
  • Sanitization: Structural integrity is ensured using Zod schemas for all input and output data.\n- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill utilizes standard package management for setup and dependency resolution, targeting verified sources.\n
  • Evidence: Documentation and scripts reference npm install and npx for installation and execution of the connector and its management tools.\n
  • Source: All package operations target the official npm registry and the author's verified package namespace.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 08:54 PM
Security Audit — agent-trust-hub — google-contacts