c-level-agents
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions intended to influence the agent's executive decision-making process.
- Ingestion points: External documents provided as input to commands such as
/c-level-agents:boardroom(e.g.,briefs/pricing-v3.md) and data captured during the/cs-onboardfounder interview process. - Boundary markers: The architecture includes 'Phase 2 isolation' during boardroom deliberation to ensure independent thinking before cross-examination, which serves as a structural deliberation boundary.
- Capability inventory: The skill generates markdown artifacts, records decisions via a
decision-logger, and orchestrates multiple persona-based agents. It mentions the use of Python tools that are restricted to the standard library. - Sanitization: There is no explicit mention of sanitization, escaping, or validation of the content within ingested briefs or interview responses to prevent embedded instruction execution.
Audit Metadata