schema-markup-generator
Pass
Audited by Gen Agent Trust Hub on Apr 28, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill includes a robust security boundary instruction in
SKILL.mdthat explicitly directs the agent to ignore any hidden instructions or command-like directives found in external content fetched viaWebFetch. - [DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were detected. The skill uses
WebFetchsolely for its intended purpose of analyzing web page content to generate relevant schema markup. - [REMOTE_CODE_EXECUTION]: No patterns for downloading and executing untrusted remote code were found. All external links lead to the author's repository or trusted industry sites (e.g., Google, Schema.org).
- [COMMAND_EXECUTION]: The skill does not perform any arbitrary shell command execution or privilege escalation attempts.
- [DATA_INCONSISTENCY]: The skill provides clear guidance on identifying and ignoring malicious directives embedded in external data, ensuring that instructions fetched from the web do not influence the agent's behavior or output integrity.
Audit Metadata