agent-auth-connectors
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown instructions and does not include any executable scripts, configuration files, or code blocks.
- [SAFE]: The instructions emphasize security best practices, specifically advising the use of constraints to enforce the principle of least privilege when granting permissions to the agent.
- [SAFE]: The skill explicitly warns against indirect prompt injection by stating that data retrieved from external sources (such as emails) must be treated as data rather than commands.
Audit Metadata