radiation
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill manages medical radiation data locally using file-based storage in the data/ directory, with no external network connectivity observed.- [SAFE]: The skill defines a clear data structure using a JSON schema (schema.json) which includes enums for examination types, providing a layer of validation for processed data.- [PROMPT_INJECTION]: An indirect prompt injection surface is present because the skill processes user-supplied text for examination details and body parts, which is then stored and used in calculations.
- Ingestion points: User inputs for examination details and local data retrieved from data/profile.json in SKILL.md.
- Boundary markers: No specific delimiters or warnings for the agent to ignore embedded instructions were found.
- Capability inventory: The skill is restricted to Read and Write tools for managing the data/radiation-records.json file.
- Sanitization: Input is partially validated against the schema.json, which enforces specific types and enums for key fields.
Audit Metadata