oauth-integrations
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions and code examples for fetching data from well-known authentication providers. Specifically, it references
api.github.comandgraph.microsoft.com, which are official and trusted API endpoints for GitHub and Microsoft services. - [INDIRECT_PROMPT_INJECTION]: The skill describes processes for ingesting data from external sources, such as user profiles and email addresses from GitHub and Microsoft APIs. This ingestion of untrusted external content constitutes a potential attack surface for indirect prompt injection.
- Ingestion points: Data returned from GitHub's
/userand/user/emailsendpoints, and Microsoft Graph's/meendpoint. - Boundary markers: The skill does not explicitly instruct the agent to use delimiters or ignore embedded instructions within the retrieved user data.
- Capability inventory: The skill utilizes the
fetchAPI to perform network operations for token exchange and data retrieval. - Sanitization: The provided examples do not include explicit sanitization or validation of the retrieved API data before it is potentially used in the agent's context.
Audit Metadata