030-architecture-non-functional-requirements
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions include a step for the AI agent to run a standard utility command in the terminal.
- Evidence: In 'Phase 0: Get Current Date', the skill directs the agent to 'run
datein the terminal to ensure accurate timestamps'. - Context: This is a low-risk operation used specifically to populate metadata fields in the resulting documentation.
- [PROMPT_INJECTION]: The skill relies on user-provided conversational input to generate technical documentation, which introduces a surface for indirect prompt injection.
- Ingestion points: Responses provided by stakeholders during the 'Conversational Information Gathering' phase.
- Boundary markers: The skill does not define specific delimiters for separating user input from the generation template in SKILL.md.
- Capability inventory: The agent has access to terminal execution for the
datecommand. - Sanitization: No explicit sanitization or validation of the architectural data is performed before document generation.
Audit Metadata