chief-of-staff
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by ingesting and processing data from potentially untrusted external sources.
- Ingestion points: The skill instructions specifically mention ingesting data from transcripts, Pull Request (PR) sources, subagent reports, and system logs to populate its internal "Wiki" (identified in
SKILL.md). - Boundary markers: No specific delimiters or instructions for the agent to ignore embedded commands are provided when processing external data.
- Capability inventory: The agent is empowered to coordinate subagents, read and write to a persistent local wiki folder, and is intended to have access to databases, browser interfaces, and critical process logs (
SKILL.md). - Sanitization: There is no evidence of content sanitization or validation routines to filter malicious instructions from the ingested content.
Audit Metadata