schema-markup
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by ingesting and processing untrusted HTML data from external sources to perform schema audits and validations.
- Ingestion points: External data is processed through the scripts/schema_validator.py script and by reading site source code.
- Boundary markers: The instructions lack explicit boundary markers or directions to the agent to ignore potentially malicious instructions hidden within the HTML comments or structured data blocks being audited.
- Capability inventory: The agent has the capability to read files and execute the local Python validator script on ingested content.
- Sanitization: While the validator script uses standard libraries for parsing, there is no logic to filter or sanitize the extracted text for hidden instructions before it is presented back to the agent for analysis.
Audit Metadata