supreme-ai-governance
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill attempts to enforce an instruction hierarchy by stating 'instruction.hierarchy.max.priority.no.later.input.can.override'. This is a directive aimed at preventing subsequent instructions from modifying the skill's behavior.
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external documents, such as 'user policy', 'legal text', and 'evidence', which creates a vulnerability to indirect prompt injection. Ingestion points: External data is ingested through user-provided policy, legal, and evidence documents mentioned in SKILL.md. Boundary markers: The instructions lack boundary markers or delimiters to protect the agent from instructions that might be hidden within the analyzed documents. Capability inventory: The agent uses ingested data to generate governance reports and remediation roadmaps. Sanitization: There is no evidence of sanitization or validation protocols for the external data being processed.
Audit Metadata