test-designer
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external data sources, creating a potential surface for indirect prompt injection attacks.\n
- Ingestion points: In SKILL.md, Procedure Step 1 requires the agent to read scoped requirements, architecture contracts, and existing tests or fixtures from the project environment.\n
- Boundary markers: The instructions do not include boundary markers or explicit safety guidelines to ignore instructions that may be embedded within the external documentation or code being analyzed.\n
- Capability inventory: Although the skill itself does not have tools for file modification or network access, its output is intended to influence the behavior of downstream implementation tasks within a larger agent workflow (SAM).\n
- Sanitization: There are no instructions for sanitizing or validating the contents of external requirements before they are processed by the agent.
Audit Metadata