grid-validator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external JSON configuration files which could potentially contain malicious instructions embedded within text fields (e.g., descriptions or titles). While the skill is restricted to read-only operations, the processing agent remains susceptible to instructions found within the audited content.
- Ingestion points: The target grid configuration file (e.g.,
body.json) provided at runtime. - Boundary markers: The skill instructions do not explicitly define delimiters or 'ignore' instructions for the sub-agent when reading the JSON body.
- Capability inventory: Access is limited to
Read,Grep, andGlobtools. - Sanitization: No explicit sanitization or validation of the input text fields is performed prior to processing.
Audit Metadata