seo-sitemap

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script sitemap_discovery.py via a vendor-provided CLI tool claude-seo. This script is used to perform automated discovery and validation of sitemaps for a given URL.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes XML sitemap data fetched from external, potentially untrusted web servers, which represents an indirect prompt injection attack surface.\n
  • Ingestion points: XML content is retrieved from external URLs provided as arguments to the analysis mode.\n
  • Boundary markers: While the skill specifies XML format validation and SSRF-safe fetching, it lacks explicit instructions for the agent to treat the ingested sitemap content as untrusted data or use delimiters.\n
  • Capability inventory: The skill can execute local scripts, read template files from a relative path (../seo-plan/assets/), and write files to the local file system (sitemap.xml, STRUCTURE.md, VALIDATION-REPORT.md).\n
  • Sanitization: The skill documentation notes that it performs XML parsing validation and checks for HTTP 200 responses.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 01:53 AM
Security Audit — agent-trust-hub — seo-sitemap