Connect external AI agents, Cursor IDEs, and Claude Desktop directly to the ThinkForge Upstash Vector and GraphRAG substrate. Powered by the open Model Context Protocol.
{
"mcpServers": {
"tresslers-matrix": {
"command": "npx",
"args": [
"-y",
"@tresslers/matrix-mcp-client"
]
}
}
}Executes a semantic vector similarity search across the entire ThinkForge intelligence substrate. Returns relevant snippets and strategic conviction metadata.
Traverses the multi-hop conceptual knowledge graph across entities. Free preview available for direct 1-hop adjacencies. Deeper reasoning requires x402 settlement.
Reads pure decrypted markdown contents of a specific intelligence dossier. Without payment proof, returns a high-fidelity strategic excerpt and payment challenge.
Returns a JSON list of all available intelligence dossier slugs across the ThinkForge matrix.
Queries intelligence updates and conviction deltas attested since a specific UTC timestamp.
Ask a natural-language question and receive structured intelligence context retrieved from Tresslers dossiers via RAG. Returns relevant chunks, source citations, conviction metadata, and graph neighborhoods. The calling LLM synthesizes the answer.
ThinkForge operates an autonomous cryptographic micro-economy. Tier 0 semantic queries and 1-hop GraphRAG adjacencies are served free. Premium full-decryption dossier reads enforce automated Base L2 USDC settlement via headers.
This gateway is structured for public listing on autonomous MCP registries. Connect your agent to start querying our RAG database directly from:
* Note: AI agents must pass valid Base L2 USDC transaction hashes in their JSON-RPC arguments to access gated tools (like read_dossier).