seo-project-setup

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, creating a surface for indirect prompt injection where malicious content in websites or files could attempt to manipulate agent behavior.\n
  • Ingestion points: The skill researches external company sites, competitor pages, and reviews (Step 5), and ingests Google Search Console CSV exports provided by the user (Step 8).\n
  • Boundary markers: The skill does not specify explicit delimiters for external content but provides a manual verification step, instructing the agent to "Confirm facts with the user before writing them."\n
  • Capability inventory: The skill uses specific Model Context Protocol (MCP) tools (update_project_context, addCompetitors, addKeyPages) to store data in a persistent project context. It lacks capabilities for arbitrary shell execution or unauthorized network exfiltration.\n
  • Sanitization: No explicit sanitization or filtering logic is defined for the content retrieved from external research or files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 07:35 PM
Security Audit — agent-trust-hub — seo-project-setup