business-investment-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest business information from a local file, which represents a potential attack surface if the file contains adversarial instructions.
- Ingestion points: The agent is prompted to read
company-context.mdto gather business context. - Boundary markers: There are no explicit instructions to use delimiters or ignore embedded prompts within the context file.
- Capability inventory: The skill is limited to performing mathematical analysis (ROI, NPV, IRR) and providing text-based recommendations; it does not utilize shell tools, file-writing capabilities, or network operations.
- Sanitization: No sanitization or verification of the context file content is performed by the skill instructions.
Audit Metadata