seo-mcp-tools-expert
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists exclusively of markdown documentation and tool usage guides. It does not include any executable scripts, binaries, or configuration files that could execute malicious actions. All analyzed components are instructional and consistent with their stated purpose.
- [DATA_EXFILTRATION]: Analysis confirmed that no credentials, tokens, or sensitive system paths are hardcoded. The skill follows security best practices by directing users to store API keys for external services (Moz, DataForSEO, Google Search Console) in environment variables rather than hardcoding them within the skill content.
- [INDIRECT_PROMPT_INJECTION]: The skill guides the agent in using tools that fetch and process external web content.
- Ingestion points: Tool outputs from
analyze_pageandcrawl_sitewhich ingest content from user-provided URLs (SKILL.md, ANALYSIS_GUIDE.md). - Boundary markers: None explicitly defined in the documentation; the skill assumes standard tool-level parsing.
- Capability inventory: The agent can perform web analysis, performance audits, and keyword research using the designated MCP tools.
- Sanitization: Not present in the instruction set; the skill focuses on the orchestration of analytical tools.
Audit Metadata