team-builder
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by design, as it reads local markdown files to define the identity and rules for subagents. Since this is the primary functionality and targets local user-controlled files, the risk is negligible.
- Ingestion points: Reads markdown files from
./agents/and~/.claude/agents/(SKILL.md). - Boundary markers: Persona content and the task description are concatenated with simple newlines.
- Capability inventory: Uses the Agent tool to spawn subagents.
- Sanitization: No validation or escaping is applied to the content of the persona files.
Audit Metadata