About Streamkap
Streamkap is a hosted MCP server and CLI for managing real-time change data capture pipelines through an AI agent. It supports pipeline creation, health checks, latency diagnosis, and dead-letter queue triage, with calls scoped to a selected Streamkap tenant.
Key features
- 01Create and manage CDC pipelines
- 02Check pipeline health and errors
- 03Investigate end-to-end latency
- 04Triage dead-letter queues
- 05Tenant-scoped account access
Use cases
- 01Connect a Postgres table to Snowflake
- 02Find failing pipelines and recent errors
- 03Investigate a production latency spike
- 04Group recent dead-letter messages by error
MCP support
The hosted MCP server at `https://mcp.streamkap.com/mcp` exposes Streamkap pipeline operations to compatible clients. OAuth uses the selected tenant; Project Keys are available for direct Kafka access and headless or CI use.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.streamkap.com/mcpclaude mcp add --transport http com-streamkap-tools https://mcp.streamkap.com/mcp{"mcpServers":{"com-streamkap-tools":{"type":"http","url":"https://mcp.streamkap.com/mcp"}}}Open in Cursor{"servers":{"com-streamkap-tools":{"type":"http","url":"https://mcp.streamkap.com/mcp"}}}Open in VS Code{"mcpServers":{"com-streamkap-tools":{"type":"http","url":"https://mcp.streamkap.com/mcp"}}}{"mcpServers":{"com-streamkap-tools":{"serverUrl":"https://mcp.streamkap.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEStreamkap MCP
Build, monitor, and troubleshoot real-time CDC pipelines from your AI agent.
Streamkap is the developer platform for real-time data. It uses change data capture (CDC) to stream changes from your databases to apps, warehouses, and AI agents the moment they happen. The hosted Streamkap MCP server lets your agent operate all of it in natural language:
- Check pipeline health: "Which pipelines are failing, and what are their latest errors?"
- Build a pipeline: "Connect my Postgres orders table to Snowflake with incremental updates."
- Diagnose latency: "End-to-end lag on prod-events jumped this morning. What changed?"
- Triage dead-letter queues: "Find every DLQ with messages in the last hour, group by error class, and suggest fixes."
This repository holds the plugin and extension manifests for each AI client, plus the Streamkap agent skill. The server itself is hosted at https://mcp.streamkap.com/mcp: there is nothing to run, and you sign in with your Streamkap account on first use.
Install
Claude
Claude.ai and Claude Desktop: add Streamkap from Settings → Connectors.
Claude Code:
/plugin marketplace add streamkap-com/streamkap-mcp
/plugin install streamkap@streamkap-mcp
Codex
codex plugin marketplace add streamkap-com/streamkap-mcp
codex plugin add streamkap@streamkap-mcp
Cursor
VS Code
Gemini CLI
gemini extensions install https://github.com/streamkap-com/streamkap-mcp
Kiro
Powers → Add Custom Power → Import from GitHub, then enter https://github.com/streamkap-com/streamkap-mcp.
Agent skill only
npx skills add streamkap-com/streamkap-mcp
Any other MCP client
{
"mcpServers": {
"streamkap": {
"type": "http",
"url": "https://mcp.streamkap.com/mcp"
}
}
}
Authentication
By default your client signs you in through OAuth with your Streamkap account, and every call is scoped to the Streamkap tenant you choose. Destructive operations (deletes and resets) are off by default for OAuth sessions.
For direct Kafka produce and consume, or for headless agents and CI, connect with a Project Key sent in the X-Streamkap-Project-Key header instead.
What's in this repository
| Path | For |
|---|---|
plugin.json, mcp.json | Agent Plugins clients: Kiro, VS Code, Codex |
.claude-plugin/, .mcp.json | Claude Code plugin and marketplace |
.codex-plugin/ | Codex and the ChatGPT plugin directory |
.cursor-plugin/ | Cursor Marketplace |
gemini-extension.json, GEMINI.md | Gemini CLI extension |
skills/streamkap/ | The Streamkap agent skill |
The skill and the version number come from the @streamkap/tools npm package, and a daily workflow opens a pull request when it publishes a new version.
Links
- Documentation: docs.streamkap.com/mcp-server
- Support: [email protected] · streamkap.com/contact-us
- Privacy Policy · Terms of Service
License
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.