unified-notifications-ops
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is purely instructional and does not include any scripts, binary files, or executable shell commands.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing external notification data, which constitutes an indirect prompt injection surface.
- Ingestion points: Notification data from GitHub, Linear, and local hooks as described in the 'Workflow' and 'Preferred Surface' sections.
- Boundary markers: The instructions do not define specific markers or delimiters to isolate notification content from the agent's control logic.
- Capability inventory: The skill utilizes external tools (MCPs, connectors, and other agent skills) to perform routing and triage actions based on the ingested data.
- Sanitization: No explicit sanitization or filtering mechanisms for the external data are specified.
- [SAFE]: The skill explicitly mandates that the agent must never expose tokens, secrets, webhook secrets, or internal identifiers, which follows security best practices.
Audit Metadata