seo-content

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it is designed to ingest and analyze external content via a user-provided URL. This is a functional requirement for SEO analysis.
  • Ingestion points: The [url] argument in SKILL.md allows the agent to fetch untrusted content from the web.
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to ignore instructions embedded within the fetched content.
  • Capability inventory: The skill utilizes DataForSEO MCP tools for data retrieval and content analysis.
  • Sanitization: The instructions do not define sanitization or filtering logic for the processed content.
  • [SAFE]: All instructional content, metrics, and E-E-A-T framework references are standard SEO best practices.
  • [SAFE]: The skill references internal documentation (skills/seo/references/eeat-framework.md) and integrates with legitimate third-party SEO data providers via the Model Context Protocol (MCP).
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 08:13 PM
Security Audit — agent-trust-hub — seo-content