open-design-new
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user requests to generate design briefs and create files within the project workspace.
- Ingestion points: User prompts and conversational input are ingested to drive the
prepare_open_design_briefandregister_open_design_artifacttools (SKILL.md). - Boundary markers: The instructions lack explicit markers or instructions to the model to treat user-provided text as data rather than instructions when passing them to downstream tools.
- Capability inventory: The agent context allows for writing files to the local file system and making specific tool calls to the Open Design framework.
- Sanitization: No sanitization, validation, or escaping rules are defined for the user-supplied brief text before it is interpolated into tool parameters or file content.
Audit Metadata