designing-ui-for-cli
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting and processing data from project-level markdown files (PRODUCT.md, DESIGN.md, DESIGN.TUI.md) to guide the agent's behavior during the design and audit process.
- Ingestion points: The
scripts/load-design-context.pyscript systematically searches and reads content from specific files in the project root,context/, ordocs/directories. - Boundary markers: While the skill employs a structured 7-stage workflow and specific reference schemas to manage context, it lacks explicit prompt delimiters or instructions to treat ingested data as untrusted, which could lead to the agent following malicious instructions embedded in project documentation.
- Capability inventory: The skill enables the agent to perform file writes (for product interviews), execute local Python scripts, and conduct design audits based on external input.
- Sanitization: No evidence of input validation, filtering, or escaping is present in the loader script or the instructional logic before the ingested content is interpolated into the agent's context.
Audit Metadata