Streamnative MCP Server
Apache Kafka and Pulsar messaging systems interaction for AI agents.
Data last scanned today · Reviewed today
Overview
The StreamNative MCP Server provides a standardized interface for AI models to communicate with Apache Kafka, Apache Pulsar, and StreamNative Cloud resources. It enables AI agents to perform administrative operations like managing topics, partitions, and schemas, alongside executing producer and consumer actions directly through the Model Context Protocol.
Our verdict
This server carries a high risk level with a security score of 40/100. Because it is community-maintained rather than official and lacks an editorially reviewed authentication method, users should exercise caution before deploying it in sensitive messaging environments. It is best suited for experimental use cases where the risk of unverified administrative access is acceptable.
- Supports multiple MCP protocol versions for broad client compatibility
- Provides native integration for Apache Kafka and Pulsar resources
- Zero direct dependencies reduces overall supply-chain complexity
- Lacks an official security review of authentication and read-only modes
- Maintained by the community rather than the official vendor
Tools
| Tool | Description | Risk |
|---|---|---|
| kafka_admin_topics_read | Reads Kafka topic information. | low |
| kafka_admin_topics_write | Modifies or creates Kafka topics. | medium |
| kafka_client_consume | Consumes messages from Kafka topics. | low |
| kafka_client_produce | Produces messages to Kafka topics. | medium |
| kafka_admin_partitions_read | Reads Kafka partition information. | low |
| kafka_admin_partitions_write | Modifies Kafka partitions. | medium |
| kafka_admin_groups_read | Reads Kafka consumer group information. | low |
| kafka_admin_groups_write | Modifies Kafka consumer groups. | medium |
| kafka_admin_schema_registry_read | Reads from the Kafka Schema Registry. | low |
| kafka_admin_schema_registry_write | Writes to the Kafka Schema Registry. | medium |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf | ||||
| ChatGPT |
Frequently asked questions
›What authentication is required to access StreamNative Cloud?
You must have a StreamNative Cloud Organization, a cluster, and a Service Account with an admin role. You will also need to download and provide the Service Account Key file when starting the server.
›Can I use this server with Kafka or Pulsar clusters outside of StreamNative Cloud?
Yes, the server supports direct connections to external Apache Kafka and Apache Pulsar clusters using specific configuration flags like --use-external-kafka or --use-external-pulsar.
›What are the limitations of providing pulsar-instance and pulsar-cluster flags?
If you provide these flags during startup, context mutation tools will be disabled.
›Are Kafka Connect operations fully supported on all platforms?
No, Kafka Connect operations are currently only tested and verified for use on StreamNative Cloud.
›Which MCP protocol versions are supported by this server?
The server supports versions 2025-11-25, 2025-06-18, 2025-03-26, and 2024-11-05, with 2025-11-25 set as the default preference.
›How can I run this server in a Docker container?
You can pull the image from streamnative/snmcp and run it using either the stdio or sse commands, ensuring you pass necessary environment variables and mount your configuration files.
Alternatives
Changelog
- SecurityStreamNative MCP Server 0.2.0
2026-08-13
- NewStreamNative MCP Server 0.1.16-rc.6
2026-04-13
- ImprovedStreamNative MCP Server 0.1.16-rc.5
2026-04-09
- FixStreamNative MCP Server 0.1.16-rc.4
2026-04-04
- ImprovedStreamNative MCP Server 0.1.16-rc.3
2026-03-12
- NewStreamNative MCP Server 0.1.16-rc.2
2026-02-05
- NewStreamNative MCP Server 0.1.16-rc.1
2026-02-03
- Improvedsnmcp-v0.1.15
2026-01-28
- NewStreamNative MCP Server 0.1.15
2026-01-27
- NewStreamNative MCP Server 0.1.15-rc.3
2026-01-27
Badge
Maintain this server? Add the live badge to your README.