seo-ecommerce
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted HTML and product data from external URLs through the page rendering and parsing workflow.
- Ingestion points: HTML, metadata, and product descriptions from user-provided URLs in SKILL.md.
- Boundary markers: None identified in the provided instructions.
- Capability inventory: Execution of local analysis scripts via the claude-seo tool and network access to marketplace APIs.
- Sanitization: The skill relies on automated parsing scripts (parse_html.py) without explicit content filtering or sanitization steps described in the markdown.
- [COMMAND_EXECUTION]: The skill executes local Python scripts (render_page.py, parse_html.py, dataforseo_merchant.py, ucp_check.py) using a tool-based runner to perform SEO audits, data normalization, and cost management.
- [EXTERNAL_DOWNLOADS]: The skill fetches marketplace data from the DataForSEO Merchant API and performs network reachability probes (HEAD requests) against external domain endpoints to validate Universal Commerce Protocol compliance.
Audit Metadata