google-calendar
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security vulnerabilities were identified in the analyzed files.
- [DATA_EXFILTRATION]: The skill reads OAuth tokens from a specific local path (/home/.z/google-oauth/token.json). This is a documented requirement for the skill to function and is used solely for authenticating with official Google APIs.
- [EXTERNAL_DOWNLOADS]: The skill relies on another local skill (google-direct-oauth) for its authentication logic, importing code from its installation directory. This is a standard cross-skill dependency and does not involve untrusted remote sources.
- [PROMPT_INJECTION]: The skill retrieves calendar event data, which could contain instructions from external sources (indirect prompt injection). However, the agent instructions provide clear context for how to handle this data, and the Python script does not execute any part of the retrieved data.
Audit Metadata