leiloeiro-ia
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation provides instructions and examples for executing local Python scripts (e.g., scan_registry.py, match_skills.py) to manage and coordinate skills within its ecosystem. These instructions include references to absolute local system paths.- [PROMPT_INJECTION]: The skill is designed to ingest and process external auction edicts and property descriptions, which constitutes a potential surface for indirect prompt injection.
- Ingestion points: The agent is instructed to read and analyze auction lot data and edicts provided by users in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters or specific warnings to ignore embedded instructions within user-provided documents.
- Capability inventory: The skill has the capability to write to the local filesystem via its governance script and execute local CLI commands.
- Sanitization: No explicit sanitization or input filtering logic for external document content is defined.- [SAFE]: The skill includes a dedicated governance module (scripts/governance.py) that implements security best practices such as action logging, rate limiting, and confirmation requests for high-risk assessments. It avoids external network dependencies and hardcoded credentials.
Audit Metadata