12-risk-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides structured workflows and extensive reference data for enterprise risk management, including specific guidance for identifying technical, regulatory, and operational risks in the Ugandan market context.\n- [SAFE]: Capability and permission boundaries are clearly defined, restricting the skill to read-only analysis and drafting of plan sections. It lacks tools for executing commands, writing to the filesystem, or performing network operations.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it is designed to analyze external business plans and assumptions.\n
- Ingestion points: External core business models, funding logic, and operating assumptions are ingested as described in the workflow sections of SKILL.md.\n
- Boundary markers: The skill includes Quality Bar and Anti-Patterns sections to enforce honest assessment and logical consistency, providing a non-technical layer of protection against accidental obedience to instructions embedded in input data.\n
- Capability inventory: Analysis is limited to producing text-based risk registers and stress tests; the skill does not have access to subprocess calls or network writing tools.\n
- Sanitization: There are no explicit instructions for sanitizing or escaping the content of ingested business plans before interpolation into the agent's context.
Audit Metadata