seo-schema
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted content from external web pages (JSON-LD, Microdata, and RDFa).
- Ingestion points: External page source code processed during detection and validation steps in
SKILL.md. - Boundary markers: The instructions lack specific delimiters or "ignore instructions" directives to prevent the agent from accidentally following commands embedded in the analyzed markup.
- Capability inventory: The skill analyzes content, validates schema types, and generates local files (
SCHEMA-REPORT.mdandgenerated-schema.json). It does not perform privileged operations or external network writes. - Sanitization: There are no instructions for sanitizing the ingested HTML/JSON content before analysis.
Audit Metadata