QueryWeaver
Let your agent turn plain-English questions into SQL using graph-backed schema context from FalkorDB QueryWeaver.
Overview
QueryWeaver is a MCP server for the Build phase that converts natural language questions into SQL using FalkorDB graph-based schema understanding.
What is this MCP server?
- Text2SQL MCP server with graph schema understanding (FalkorDB)
- OCI package docker.io/falkordb/queryweaver:0.0.11
- Streamable-http transport for local or containerized /mcp endpoint
- Open-source repo at github.com/FalkorDB/QueryWeaver
- Registry version 0.0.11 on MCP server schema 2025-09-29
- 1 OCI package identifier docker.io/falkordb/queryweaver:0.0.11
- Transport type streamable-http
Community signal: 1k GitHub stars.
What problem does it solve?
Your agent keeps hallucinating table joins because flat schema dumps do not capture how your data model really connects.
Who is it for?
Backend-focused solos building agent-assisted analytics, admin tools, or RAG-adjacent data features on FalkorDB or graph-informed SQL workflows.
Skip if: Non-technical founders who only need a spreadsheet, or production systems without human or automated review of generated SQL.
What do I get? / Deliverables
After you run QueryWeaver and register its MCP endpoint, the agent can generate schema-aware SQL from conversational prompts against your connected database context.
- MCP tools for natural-language SQL generation
- Graph-informed schema context for agent queries
- Local or private deployment via container transport
Recommended MCP Servers
Journey fit
Natural-language SQL access is wired while you build data features and hook agents to real stores—not a launch or growth marketing tool. Integrations is the right shelf because QueryWeaver is an MCP bridge between the LLM and your database schema, not a UI or deploy step.
How it compares
Self-hosted Text2SQL MCP over graph schema, not a managed warehouse BI connector or a prompt-only SQL skill.
Common Questions / FAQ
Who is QueryWeaver for?
Developers shipping data-backed products who want MCP-enabled agents to draft SQL with graph schema context from the FalkorDB QueryWeaver server.
When should I use QueryWeaver?
Use it during Build integrations when you connect Claude Code or Cursor to a database and need Text2SQL tools instead of hand-writing every exploratory query.
How do I add QueryWeaver to my agent?
Pull docker.io/falkordb/queryweaver:0.0.11, expose the streamable-http MCP URL (documented as https://localhost:5000/mcp for local runs), and register that remote in your MCP client.