seosiri-keyword-cluster-mcp 1.0.0

A sovereign, local-first MCP server for semantic keyword clustering, search intent classification, cannibalization detection, and topical authority mapping.

seosiri-keyword-cluster-mcp

📖 Official Architecture & Documentation: SEOSiri Keyword Clustering Guide | Central MCP Directory

An open-source, local-first Model Context Protocol (MCP) server for Semantic Keyword Clustering, Search Intent Classification, Cannibalization Detection, and Topical Authority Mapping.

💖 Sponsorship, B2B Custom Solutions & Attribution

👨‍💻 Lead Architect & Attribution

Designed and engineered by Momenul Ahmad, Lead Architect and Founder of SEOSiri.

🚀 Capabilities & Tools (10 Production Tools)

• cluster_keywords_by_similarity: Groups keyword lists into semantic clusters using token overlap.
• classify_search_intent: Categorizes queries into Informational, Navigational, Commercial, or Transactional intent.
• detect_keyword_cannibalization: Identifies conflicting URLs targeting the same search query.
• generate_topical_authority_map: Compiles pillar-cluster content hierarchy trees.
• calculate_keyword_difficulty_score: Computes algorithmic query difficulty scores ($0\text{--}100$).
• extract_lsi_semantic_variants: Identifies Latent Semantic Indexing terms and entity phrases.
• export_cluster_parquet_buffer: Formats keyword clusters into columnar Parquet buffers for DuckDB/S3.
• sanitize_keyword_payload: Strips scripts and malformed tags from inputs.
• get_live_keyword_throughput_metrics: Monitors system health and processing latency.
• get_keyword_server_specifications: Returns protocol specs and capability matrices.

Quickstart

pip install -e .
pytest tests/test_keyword_cluster.py

🔌 Claude Desktop / Cursor Configuration

Add this block to your claude_desktop_config.json:

{
  "mcpServers": {
    "seosiri-keyword-cluster": {
      "command": "uv",
      "args": [
        "run",
        "--github",
        "SEOSiri-Official/keyword-cluster-mcp",
        "src/main_server.py"
      ]
    }
  }
}

License

Distributed under the MIT License.