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 is designed to ingest and analyze external HTML content, which serves as a surface for indirect prompt injection.
- Ingestion points: Processes HTML files or content provided by the user through instructions in SKILL.md and the
scripts/schema_validator.pyscript. - Boundary markers: The instructions do not explicitly define delimiters for external content, relying on the agent's default handling of provided data.
- Capability inventory: The included Python script is limited to local file reading and static JSON-LD parsing. It does not contain network request, file writing, or arbitrary command execution capabilities.
- Sanitization: The validation script uses standard library parsers (
html.parserandjson.loads) to extract and validate specific JSON structures, which limits the risk of processing unintended instructions embedded in HTML.
Audit Metadata