mino-reproducible-development
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process external artifacts and outputs from multiple agents.
- Ingestion points: The workflow involves reading 'raw materials, real code, and tests' (SKILL.md) and 'raw requests' (references/reproduction.md) to integrate them into a trace.
- Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore potential instructions embedded within the processed artifacts.
- Capability inventory: The skill is capable of routing tasks to other agents ($mino-*) and executing 'solvers' within workspaces (references/reproduction.md).
- Sanitization: The skill does not define sanitization or filtering protocols for external content before it is processed or evaluated by the agent.
Audit Metadata