diagnose-growth
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection when processing external, untrusted data.
- Ingestion points: Step 1 in
SKILL.mdinstructs the agent to read 'customer evidence' and 'recent reviews'. - Boundary markers: The instructions lack delimiters or specific warnings to ignore instructions that might be embedded in user-generated content or reviews.
- Capability inventory: The skill's primary capabilities are reading local files and writing to a backlog file (
.business/backlog.md). It has no network access or shell execution tools. - Sanitization: There is no evidence of sanitization or filtering of the external evidence before it is analyzed by the agent.
Audit Metadata