seo-audit
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to crawl and analyze content from up to 500 external web pages, creating a significant attack surface.
- Ingestion points: Website HTML, metadata, and content are retrieved via
scripts/fetch_page.pyand processed by the agent and its specialized subagents (SKILL.md). - Boundary markers: The instructions lack explicit delimiters or warnings to the agent to ignore instructions embedded within the crawled HTML content.
- Capability inventory: The skill executes local scripts, generates file-based reports, and interacts with external APIs and browser automation tools (Playwright).
- Sanitization: There is no evidence of sanitization, filtering, or validation of the retrieved web content before it is processed by the LLM.
- [COMMAND_EXECUTION]: The skill process involves the execution of Python scripts (e.g.,
scripts/fetch_page.py) to perform the crawling operations described in the process section. - [EXTERNAL_DOWNLOADS]: The skill references the use of Playwright for visual analysis and screenshots, which typically requires the installation of browser binaries, and integrates with the DataForSEO service for live SERP and backlink data.
Audit Metadata