data-ai MCP Server
Open-source deep research MCP. Qwen3-30B-A3B-Thinking via OpenRouter, cited web synthesis for Claude Code, Codex, Cursor, Hermes and any MCP-compatible agent.
Discovered via github-topic:claude-mcp and last synced 2mo ago.
Install instructions not detected yet
Check the source repository for the latest setup steps.
Full-stack smoke
Wire gigaxity into Claude Code
Role
LLM endpoint
Companion MCPs (Triple Stack)
Primary search source
Routing skill + agent instructions
Doc
1–2 s factual answer with citations
What you do
Should route to
Core install
Research basis
Deep synthesis with optional CoT depth control over pre-gathered content
Citation-aware synthesis over pre-gathered content; CRAG-style quality gate, contradiction surfacing, outline-guided generation
Feature
Fast conversational answer (direct LLM call, no search hop)
Method
Purpose
List focus modes
Branch
Raw multi-source aggregation across SearXNG, Tavily, and LinkUp with RRF fusion. No LLM call.
Multi-source search only (no LLM)
Hosted Qwen3-30B-A3B-Thinking via OpenRouter
Combined pipeline: multi-source search plus LLM synthesis with citations, in one call.
Either, plus optional remote model server
Exploratory expansion — surfaces explicit, implicit, related, and contrasting angles, then flags knowledge gaps