chief-of-staff
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by coordinating subagents based on external 'context pointers' such as research notes, git commits, and logs. This creates an attack surface where malicious instructions could be embedded in those external sources to influence the agent's behavior.
- Ingestion points: Research notes, git commits, system logs, and database content specified in
SKILL.md. - Boundary markers: There are no instructions defining clear boundaries or 'ignore' delimiters for the untrusted external data.
- Capability inventory: The skill orchestrates subagents and is designed to interact with logs, databases, and web browsers.
- Sanitization: No sanitization, escaping, or validation logic is defined for the external content processed by the skill.
Audit Metadata