Apache Flink MCP Server
MCP server that lets AI assistants monitor Apache Flink clusters and jobs in plain
The repository was last updated 4 days ago and is published under MIT.
- Pricing
- Open source
- Platforms
- CLI, self-hosted, API
- GitHub stars
- 13
- Last commit
- 4 days ago
- Language
- Python
- License
- MIT
Similar MCP servers
About Apache Flink MCP Server
Apache Flink MCP Server is an open-source MCP server for developer tools. It is available on CLI, self-hosted and API.
The README says this server connects AI assistants to a running Apache Flink cluster through the Flink REST API, so questions about jobs, slots and TaskManagers can be asked conversationally instead of through the Flink Web UI. Tools cover cluster overview, job listings and details, job exceptions, job metrics, TaskManager resources and uploaded JAR files.
It supports several clusters in one server process. Each tool accepts an optional connection name, so separate Flink deployments can be registered under different names and queried independently, with the default connection used when none is given.
The project is written in Python, requires Python 3.8 or higher, and is released under the MIT License. It is aimed at developers and operators who run Flink streaming applications and use an MCP-compatible client such as Claude Desktop or Continue. The README notes that access is read-only by default and that write operations need extra configuration.
Tools this server provides (10)
Taken from the project documentation; the running server may differ.
- initialize_flink_connection
- Connect to the Flink REST API.
- get_connection_status
- Check connection status.
- get_cluster_info
- Overview of the Flink cluster.
- list_jobs
- List all Flink jobs with status.
- get_job_details
- Comprehensive job details by ID.
- get_job_exceptions
- Fetch job-level exceptions.
- get_job_metrics
- Fetch metrics for a job.
- list_taskmanagers
- List TaskManagers with resources.
- list_jar_files
- List uploaded JAR files.
- send_mail
- Send an email notification.
What it does
- Cluster overview monitoring
- Returns an overview of the Flink cluster, including jobs, slots and TaskManagers.
- Job status and details
- Lists current and recent jobs with their status and fetches detailed information for a given job ID.
- Exception retrieval
- Fetches exceptions recorded for a specific job to help with debugging failures.
- Job and cluster metrics
- Retrieves selected metrics for a running Flink job.
- Multi-cluster connections
- Registers several Flink clusters under named connections and targets each one separately.
- TaskManager and JAR inspection
- Lists registered TaskManagers with their resources and uploaded JAR files.
Good for
- Checking the health and slot usage of a Flink cluster from an AI chat
- Pulling exception details for a failing job to speed up troubleshooting
- Comparing job status across separate Flink clusters from one assistant
- Reviewing TaskManager resource allocation during performance investigations
Pricing
Open source
Free and open source under the MIT License.
Apache Flink MCP Server FAQ
Which clients can use this server?
The README lists MCP-compatible clients such as Claude Desktop and Continue, and shows a Continue.dev configuration using streamable HTTP.
Does it need access to a Flink cluster?
Yes. The README lists a running, accessible Apache Flink cluster as a prerequisite and notes that the Flink REST API must be enabled.
Can it manage more than one cluster?
Yes. Connections are registered with a name, and each tool accepts an optional connection argument to choose the target cluster.
Is it read-only?
The README says it provides read-only access to cluster information by default, and that write operations need additional configuration and security considerations.
Is Apache Flink MCP Server free?
Yes. Apache Flink MCP Server is open source, so you can use and self-host it for free.
What are the best alternatives to Apache Flink MCP Server?
The closest alternatives to Apache Flink MCP Server are Northflank MCP by usefulapi, Airflow MCP Server, MCP, awesome-mcp-servers and Weblate MCP by usefulapi.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.
