ito-data-atlas-agent
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an architecture designed to ingest data from untrusted external sources, including the public web, X (Twitter), GitHub, and various venue documents.
- Ingestion points: Data collection from public web, X, GitHub, and API metadata as defined in the 'Architecture Pattern' section of SKILL.md.
- Boundary markers: The skill does not define specific syntax delimiters but mandates a 'Risk reviewer' lane to validate outputs.
- Capability inventory: The skill itself is purely architectural and does not contain executable code, subprocess calls, or network operations.
- Sanitization: The architecture explicitly includes a 'Risk reviewer' role responsible for checking 'prompt-injection exposure' and 'resolution ambiguity' before any human review or potential execution.
Audit Metadata