seo-schema-jsonld
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of HTML and DOM snapshots to extract and generate schema. This ingestion of arbitrary web content presents a surface for indirect prompt injection if a website contains malicious instructions.
- Ingestion points: The skill reads from rendered_dom and raw_html variables.
- Boundary markers: No explicit delimiters or boundary markers are mentioned to isolate the external content from the agent's instructions.
- Capability inventory: The skill has access to Bash for script execution, WebFetch for network operations, and file system tools like Read and Glob.
- Sanitization: The skill includes parsing logic, validation against Tier-1 templates, and specific instructions to the agent to avoid inventing data, which mitigates the risk.
- [COMMAND_EXECUTION]: The skill executes a local Node.js script as part of its verification process to ensure JSON-LD validity.
- Evidence: The verification step uses the command: node ${CLAUDE_SKILL_DIR}/../../scripts/validate-jsonld.mjs --url .
Audit Metadata