seo-keyword-insights
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external URLs (Workflow Step 1 in
SKILL.md) using thedataforseo_v3_on_page_content_parsing_livetool. This creates a surface for indirect prompt injection where instructions hidden on the target website could influence the agent's behavior. - Ingestion points: The
dataforseo_v3_on_page_content_parsing_livetool fetches content from user-supplied domains and specific pages for analysis. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore instructions' warnings when processing the parsed site content.
- Capability inventory: The skill uses
sandbase_call_toolto interact with various SEO and search engine capabilities (as listed inreferences/sandbase-api-map.md), which involves performing network requests via the provider's API. - Sanitization: The skill lacks explicit instructions to sanitize or validate the content retrieved from external pages before it is interpreted by the model.
Audit Metadata