25-ai-agent-evidence-pack-spec
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process multiple external workspace documents to generate compliance evidence specifications, creating a potential surface for instructions embedded in those documents to influence the agent.
- Ingestion points: The skill instructions in SKILL.md identify several required inputs from the workspace, including AI_Agent_Architecture_Spec.md, AI_Agent_SOC2_Control_Pack.md, and AI_Agent_Runbook.md.
- Boundary markers: The skill does not define specific delimiters or instructions to the agent to ignore or isolate instructions that might be contained within these external project files.
- Capability inventory: The skill utilizes standard file read, search, and write capabilities to generate documentation. No high-risk tools such as network exfiltration or arbitrary shell execution were identified in the instructions.
- Sanitization: No validation or sanitization logic is implemented to filter the content of ingested files before they are processed by the agent to generate output.
Audit Metadata