deliver-acceptance-criteria
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or behaviors were identified. The skill consists entirely of instructional markdown and reference templates for document generation.
- [NO_CODE]: The skill package does not contain any programming language files, shell scripts, or software dependencies. All logic is expressed in natural language instructions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied requirements and feature stories as input. 1. Ingestion points: The agent is instructed to confirm the story or feature scope by asking for user input in SKILL.md. 2. Boundary markers: The skill does not define specific delimiters or instructions to isolate user-provided data. 3. Capability inventory: There are no capabilities for code execution, tool use, filesystem access, or network operations. 4. Sanitization: No input sanitization is performed. Assessment: The risk is negligible because the agent has no dangerous capabilities to exploit through injected data.
Audit Metadata