ai-interaction-design
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The analyzed skill consists of markdown documentation and metadata configuration for designing human-AI workflows. No malicious patterns, command executions, or data exfiltration attempts were detected.
- [NO_CODE]: The skill does not include any executable scripts (.py, .js, .sh), binary files, or installation commands. It functions strictly as a set of instructional prompts and configuration templates.
- [INDIRECT_PROMPT_INJECTION]: The skill's design framework describes the ingestion of external context and source data (found in SKILL.md under 'Design input and context' and 'Expose provenance').
- Ingestion points: Instructions in SKILL.md suggest incorporating 'source data' and 'context' into the agent's workflow.
- Boundary markers: The skill does not explicitly define markers or delimiters for the data it processes as it is a documentation-only resource.
- Capability inventory: No capabilities are present; the skill lacks subprocess calls, network operations, or file-writing logic.
- Sanitization: No automated sanitization or escaping logic is provided within the instructions.
Audit Metadata