ai-system-assessment
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill evaluates untrusted data from target repositories, creating a surface for indirect prompt injection where deceptive content in the repository could potentially influence the agent's assessment logic or behavior.
- Ingestion points: Target repository files (READMEs, instruction files, and code) are read and processed using
grepandlsduring the assessment workflow as specified in SKILL.md. - Boundary markers: No explicit delimiters are defined to separate the untrusted repository content from the agent's own instructions within the prompt context.
- Capability inventory: The skill possesses capabilities to read files and write a report file (
AI_SYSTEM_ASSESSMENT.md). It is explicitly restricted from executing, installing, or modifying code within the target system (SKILL.md, Rule 1). - Sanitization: The skill does not perform explicit sanitization or escaping of the ingested repository content before it is processed by the agent.
Audit Metadata