verify-ai-department-package
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external artifacts, which is an attack surface for indirect prompt injection.
- Ingestion points: The agent is tasked with reading and auditing "AI department packets," missions, skills, manifests, evaluation fixtures, and server code (SKILL.md).
- Boundary markers: The instructions lack specific technical delimiters to separate the untrusted audit data from the agent's core instructions.
- Capability inventory: The verification process requires reading files, inspecting server code, and "exercising" adversarial and approval-bypass test cases (SKILL.md).
- Sanitization: There are no instructions for sanitizing or filtering external content to prevent the agent from executing malicious payloads embedded in the audited files.
Audit Metadata