seo-wordpress-manager
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data fetched from a WordPress site to generate SEO optimizations, creating an attack surface where malicious content on the site could influence the agent's behavior.
- Ingestion points: Data including post titles, excerpts, and full content is ingested from the WordPress GraphQL endpoint in
scripts/analyze_seo.pyand passed to the agent viaanalysis.json. - Boundary markers: The instructions for the agent to process the analysis report do not define clear delimiters or include warnings to ignore instructions potentially embedded in the fetched WordPress content.
- Capability inventory: The skill has the capability to modify site content through the
updatePostSeomutation implemented inscripts/yoast_batch_updater.py. - Sanitization:
scripts/analyze_seo.pyuses regex to strip HTML tags, but it does not filter the text for potential prompt injection payloads that could compromise the integrity of the generated SEO metadata.
Audit Metadata