seo-headings-structure
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of HTML snapshots (
rendered_domorraw_html). The ingestion of arbitrary web content combined with capabilities likeBashandWebFetchcreates a surface for indirect prompt injection, where malicious instructions hidden in a webpage could influence the agent's behavior. - Ingestion points: Untrusted HTML data enters the context via the
PageSnapshotdescribed in the Audits section ofSKILL.md. - Boundary markers: There are no explicit instructions or delimiters defined to isolate the processed HTML content or to instruct the agent to ignore any natural language instructions embedded within the data.
- Capability inventory: The skill utilizes several tools including
Bash,WebFetch,Read,Grep, andGlobacross its auditing and verification tasks. - Sanitization: The instructions do not specify any sanitization, filtering, or validation steps for the external HTML content before it is processed by the agent.
Audit Metadata