Conduit MCP
Connects AI agents to Conduit data streams for ingesting events, querying data, and managing schemas. It enables seamless integration between services and agents with real-time forwarding and automated schema detection.
README
@useconduit/mcp
MCP server for Conduit ā connect any AI agent to your data streams.
š API Docs Ā· š Website Ā· š¦ npm Ā· š» GitHub
Quick Start
Get an API key from platform.usecondu.it/tokens, then add to your AI editor config:
Claude Code / Claude Desktop
Add to ~/.claude/settings.json:
{
"mcpServers": {
"conduit": {
"command": "npx",
"args": ["-y", "@useconduit/mcp"],
"env": {
"CONDUIT_API_KEY": "conduit_sk_..."
}
}
}
}
Cursor
Add to .cursor/mcp.json in your project:
{
"mcpServers": {
"conduit": {
"command": "npx",
"args": ["-y", "@useconduit/mcp"],
"env": {
"CONDUIT_API_KEY": "conduit_sk_..."
}
}
}
}
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"conduit": {
"command": "npx",
"args": ["-y", "@useconduit/mcp"],
"env": {
"CONDUIT_API_KEY": "conduit_sk_..."
}
}
}
}
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
CONDUIT_API_KEY |
ā | ā | Your Conduit API key |
CONDUIT_API_URL |
ā | https://api.usecondu.it |
Custom API endpoint |
Tools
| Tool | Description |
|---|---|
conduit_list_streams |
List all data streams |
conduit_get_schema |
Get schema for a stream (columns, types, codecs) |
conduit_create_stream |
Create a new stream |
conduit_ingest |
Send events to a stream |
conduit_list_events |
Query events with pagination & time range filters |
conduit_add_forward |
Add a forwarding destination (HTTP, MQTT, or WebSocket) with auth options |
conduit_stream_stats |
Get ingestion statistics (event count, rate, latency) |
conduit_analyze_schema |
Analyze a JSON payload for optimal schema |
conduit_feedback |
Submit feedback to the Conduit team |
conduit_backfill |
Replay historical events to forwarding destinations |
conduit_backfill_status |
Check backfill job progress or list all jobs |
conduit_time_range |
Get earliest/latest event timestamps for a stream |
Resources
| URI | Description |
|---|---|
conduit://streams |
All streams |
conduit://streams/{name} |
Stream details + schema |
conduit://stats |
Platform-wide statistics |
API Documentation
Full interactive API docs are available at api.usecondu.it/docs (powered by Scalar).
Key endpoints:
| Method | Endpoint | Description |
|---|---|---|
POST |
/v1/{tenant}/{stream} |
Ingest events (any protocol: HTTP, WebSocket, MQTT) |
GET |
/api/v1/streams |
List streams |
GET |
/api/v1/streams/{name} |
Stream details + schema |
GET |
/api/v1/streams/{name}/events?from=&to=&limit= |
Query events with time range |
GET |
/api/v1/streams/{name}/stats |
Ingestion statistics |
GET |
/api/v1/streams/{name}/schema-history |
Schema evolution history |
POST |
/api/v1/streams/{name}/forwards |
Add forwarding destination |
POST |
/api/v1/streams/{name}/backfill |
Start backfill job |
GET |
/api/v1/streams/{name}/backfill |
List backfill jobs |
GET |
/api/v1/streams/{name}/time-range |
Event time range |
POST |
/api/v1/tokens |
Create API token |
GET |
/api/v1/account |
Account info + usage |
Machine-readable API specs:
- OpenAPI 3.1: api.usecondu.it/openapi.json
- LLM-friendly: api.usecondu.it/llms.txt
- LLM full spec: api.usecondu.it/llms-full.txt
What is Conduit?
Conduit is the lightweight data layer between your services. Send any JSON ā schemas aren't defined, they emerge. And they evolve as your sources do.
- One endpoint, any protocol ā HTTP, WebSocket, MQTT/S on the same path
- AI-powered schema detection ā zero configuration, runs locally (your data never leaves)
- Real-time forwarding ā HTTP webhooks, MQTT brokers, WebSocket endpoints with full auth options
- Backfill ā replay historical events to new destinations
- Built for agents ā MCP-native from day one, feedback loop included
- European infrastructure šŖšŗ ā all data hosted and processed in the EU, GDPR-compliant by design
How agents use Conduit
Agent: "I see live sensor data flowing in. I'll pipe it to our datawarehouse and add hot storage."
āø conduit_list_events(stream: "sensors", limit: 1000)
ā³ 1,000 events Ā· 18.4 MB raw ā 1.1 MB stored (94% compressed)
āø conduit_add_forward(stream: "sensors", type: "http", url: "https://dwh.acme.io/ingest")
ā³ Destination added Ā· forwarding to datawarehouse
āø conduit_add_forward(stream: "sensors", type: "mqtt", broker: "mqtt://redis.acme.io", topic: "hot/sensors")
ā³ Destination added Ā· 2 destinations active
āø conduit_backfill(stream: "sensors", from: "2026-02-24", to: "2026-03-02")
ā³ Backfilling 89,929 events from last 7 days
Learn more at usecondu.it
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.