seo-sitemap
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze XML sitemaps from external websites, which creates a surface for indirect prompt injection if instructions are embedded within the fetched XML data. The skill includes instructions to mitigate technical risks like XML External Entity (XXE) attacks.
- Ingestion points: External XML sitemap files fetched from remote URLs via the WebFetch tool.
- Boundary markers: Instructions direct the agent to treat fetched content as data and specify parsing constraints, though no explicit delimiters are provided for the internal prompt structure.
- Capability inventory: The agent uses tools including WebFetch, Bash, and Write to analyze site structure and generate reporting files.
- Sanitization: Instructions explicitly require disabling DTD and network resolution during XML parsing and bounding response sizes to prevent resource exhaustion.
- [DATA_EXFILTRATION]: The skill performs network operations using WebFetch to retrieve files from external domains for SEO analysis. While these operations are core to the skill's purpose, they involve initiating connections to non-whitelisted domains.
Audit Metadata