problem-framing
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill transforms untrusted user input into structured context for other agents, creating a potential attack surface.
- Ingestion points: Workflow Step 1 in
SKILL.mdgathers raw input from users. - Boundary markers: The output format defined in
SKILL.mdlacks delimiters or instructions that would tell downstream agents to treat the content as data only. - Capability inventory: The skill uses a Write tool to create the
design/01-problem-framing.mdfile. - Sanitization: There is no evidence of sanitization or escaping of the user-provided text before it is saved to the filesystem.
Audit Metadata