zr-review-discoveries
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes discovered-task reports and context commands retrieved from external sources. These inputs could contain embedded instructions intended to manipulate the agent's triage decision.
- Ingestion points: Data retrieved via 'probe discover get' in SKILL.md.
- Boundary markers: No specific delimiters or instructions to ignore embedded content are present.
- Capability inventory: Executes triage decisions via 'probe discover review' as defined in 'SKILL.md'.
- Sanitization: No data validation or sanitization steps are documented.
Audit Metadata