business-case
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted user-provided data and interpolating it into a generated markdown file.
- Ingestion points: User-supplied responses to prompts regarding business problems, strategic alignment, and financial projections described in SKILL.md.
- Boundary markers: The template does not define clear delimiters (such as XML tags or specific blocks) to isolate user-provided text from the skill's instructional structure.
- Capability inventory: The skill is designed to write output to the file system at
docs/strategic/BUSINESS_CASE.md. - Sanitization: There is no evidence of input validation or escaping mechanisms to prevent malicious instructions within user data from affecting the agent's behavior during document generation.
Audit Metadata