distill
Pass
Audited by Gen Agent Trust Hub on May 5, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection because it ingests and processes agent instruction files (
AGENTS.md,rules/*.md) which could contain adversarial content. - Ingestion points: Files located at
ai-workspace/rules/*.mdandAGENTS.md. - Boundary markers: None explicitly defined in the orchestration instructions.
- Capability inventory: Executes local sub-skills (
/imperatives,/policy-algebra,/visualize) and performs file write operations to theai-workspace/research/directory. - Sanitization: Not explicitly described; the skill relies on the underlying sub-skills to handle content safely.
Audit Metadata