flowspec-mcp
MCP server for FlowSpec that exposes project specifications to Claude Code and other AI tools.
README
flowspec-mcp
MCP (Model Context Protocol) server for FlowSpec — exposes project specifications to Claude Code and other MCP-compatible AI tools.
Quick Start
Add to your ~/.claude.json:
{
"mcpServers": {
"flowspec": {
"command": "npx",
"args": ["-y", "flowspec-mcp"],
"env": {
"FLOWSPEC_MODE": "cloud",
"DATABASE_URL": "your-neon-connection-string",
"FLOWSPEC_USER_ID": "your-clerk-user-id"
}
}
}
}
Local Mode (with FlowSpec Desktop)
{
"mcpServers": {
"flowspec": {
"command": "npx",
"args": ["-y", "flowspec-mcp"],
"env": {
"FLOWSPEC_MODE": "local"
}
}
}
}
Local mode connects to the FlowSpec desktop server at http://localhost:3456.
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
FLOWSPEC_MODE |
No | cloud |
cloud for direct Neon SQL, local for desktop server HTTP |
DATABASE_URL |
Cloud mode | — | Neon Postgres connection string |
FLOWSPEC_USER_ID |
Cloud mode | — | Clerk user ID — find yours at flowspec.app/account |
FLOWSPEC_LOCAL_URL |
No | http://localhost:3456 |
Desktop server URL (local mode) |
Available Tools (v5.6.0 — 32 tools)
Set FLOWSPEC_TOOLS=core for 11 essential tools only. Default (all) loads all 32.
Read Tools
flowspec_list_projects— List all projects with names and datesflowspec_get_json— Get full JSON spec for a project (optimised for Claude Code)flowspec_get_project— Get raw canvas_state JSONflowspec_search_nodes— Search nodes by label across all projectsflowspec_get_screen_context— Get screen/region/element structure
Write Tools
flowspec_create_project— Create a new projectflowspec_update_project— Update project name or canvas stateflowspec_delete_project— Delete a projectflowspec_clone_project— Clone a project for backup or branchingflowspec_create_node— Add a node (datapoint, component, transform, table, actor)flowspec_update_node— Update node data or positionflowspec_delete_node— Remove a node and connected edgesflowspec_create_edge— Connect two nodes with an edge typeflowspec_update_edge— Update edge type, label, or handle positionsflowspec_delete_edge— Remove an edgeflowspec_analyse_project— Run orphan node and duplicate label analysisflowspec_validate_project— Validate data flow semantics and type matching
Bulk / Import Tools
flowspec_import_json— Import a JSON spec (create/merge nodes, edges, screens)flowspec_auto_layout— Auto-arrange nodes using dagre hierarchical layout
Screen & Region Tools
flowspec_upload_image— Upload a wireframe image and get its URLflowspec_create_screen— Add a wireframe screen to a projectflowspec_update_screen— Update screen name or image propertiesflowspec_delete_screen— Delete a screen and its regionsflowspec_add_region— Add an annotated region to a screenflowspec_update_region— Update region position, size, or linked elementsflowspec_remove_region— Remove a region from a screen
Decision Tree Tools
flowspec_list_decision_trees— List decision trees for a projectflowspec_get_decision_tree— Get a decision tree with full node/edge structureflowspec_delete_decision_tree— Delete a decision treeflowspec_analyse_decision_tree— Analyse tree depth, outcomes, and issues
Boring Logic Board Tools
flowspec_get_logic_board— Get the logic board (input/output/process/decision canvas)flowspec_upsert_logic_board— Save the logic board state for a project
Development
npm install
npm run build
node dist/index.js
Notes
- MCP SDK pinned to
1.12.1due to zod v4 compatibility constraints in later versions - Node.js >= 18.0.0 required
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.