content-publishing
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests markdown or HTML content generated by other agents to publish to a WordPress site. If the source content contains malicious instructions, the agent may be influenced to perform unintended actions during the formatting or verification stages.
- Ingestion points: Input field "Content" described in SKILL.md.
- Boundary markers: No specific delimiters or "ignore instructions" markers are defined for the content input.
- Capability inventory: WordPress REST API (POST), Google Search Console service account access, and network verification of published URLs.
- Sanitization: The skill mentions formatting markdown to HTML but lacks explicit sanitization or filtering of the input content to remove embedded scripts or hidden instructions.
- [NO_CODE]: The skill consists exclusively of markdown documentation and configuration metadata without any accompanying scripts or binary executables.
Audit Metadata