component-validator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill audits Datex Studio component files provided as input, which creates a surface for indirect prompt injection if the component data contains instructions aimed at the agent.\n
- Ingestion points: The sub-agent defined in
SKILL.mdreads component files, and thescripts/validate-component.pyscript parses these files to check for metadata constraints.\n - Boundary markers: The sub-agent's prompt template does not include boundary markers or specific instructions to ignore embedded commands in the processed data.\n
- Capability inventory: The sub-agent is equipped with
Read,Grep, andGlobtools for file inspection. Thevalidate-component.pyscript performs file reads and JSON parsing within the user's environment.\n - Sanitization: No sanitization of the component file content is performed prior to the audit analysis.\n- [COMMAND_EXECUTION]: The skill distributes a Python utility (
scripts/validate-component.py) and provides instructions for its configuration as a local shell-executed hook in the agent environment.
Audit Metadata