rustfs-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@rustfs-mcpShow me the objects in my bucket named data"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
rustfs-mcp
A per-add-on RustFS (S3) object-storage admin MCP for the kubero chat. When a
user selects a RustFS add-on in the chat's @-context selector, the kubero server
(the feature-MCP broker) surfaces this server's tools namespaced s3__*, gated by
the add-on selection. The 5th feature MCP after postgres/redis/mysql/kafka.
RustFS is an S3-compatible object store (operator CR kind Tenant, group
rustfs.com); in-cluster its S3 API is http://<instance>-io:9000, plain HTTP
with path-style addressing.
Architecture
ONE long-lived ClusterIP Deployment (rustfs-mcp:3010 in ns kubero-mcp) serves
every RustFS add-on in the workload cluster. The target endpoint + credentials
access mode arrive per request as HTTP headers set by the kubero broker:
Header | Value |
|
|
|
|
--connection-from-request reads these via the FastMCP lowlevel request_ctx
contextvar; a missing/invalid access mode ⇒ restricted (fail-safe). The kubero
server reaches the service via apiserver port-forward — no public ingress.
Related MCP server: MinIO MCP Server
Tools
Reads (always available): list_buckets, list_objects, stat_object,
bucket_usage, get_object_text, get_bucket_policy.
Writes (refused under restricted mode): create_bucket, delete_bucket,
put_object_text, delete_object, set_bucket_policy.
Run
pip install -r requirements.txt
# Production (in-cluster): per-request connection from headers.
python main.py --transport streamable-http --connection-from-request \
--streamable-http-host 0.0.0.0 --streamable-http-port 3010
# Local (stdio): binds one endpoint from S3_ENDPOINT_URL / AWS_* env (.env.template).
python main.pyCI / deploy
Mirrors the other stackblaze feature MCPs: auto-tag-on-main → docker-release
(amd64, self-hosted kubero runner, ghcr.io/stackblaze/rustfs-mcp) → deploy.yaml
rolls the ClusterIP Deployment to the workload cluster (gated on
KUBECONFIG_B64_SHARED). Repo secrets: CHAIN_PAT, KUBECONFIG_B64_SHARED. Seed a
v0.1.0-sb.0 tag once so auto-tag can increment.
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for Gainium — manage trading bots, deals, and balances via AI assistants
MCP server for Sendbird — chat users, channels, members, and messages from your AI client.
Cloud-hosted MCP server for durable AI memory
An MCP server that provides read access to your cloud storage providers, bank accounts and more.
Related MCP Servers
- FlicenseAqualityDmaintenanceProvides tools for interacting with MinIO and S3-compatible object storage through MCP clients like Claude. It enables comprehensive bucket and object management, including listing, creating, uploading, and generating presigned URLs.132-
- AlicenseNot gradedqualityDmaintenanceIt Integration for MinIO / S3-compatible object storage, providing AI assistants with direct access to bucket management, object CRUD, presigned URLs, policies, lifecycle rules, and storage analytics.1MIT
- FlicenseNot gradedqualityDmaintenanceA Model Context Protocol (MCP) server that provides a bridge between MCP-compatible clients and MinIO object storage. It exposes MinIO operations as MCP tools for seamless bucket management and object operations.4-
- AlicenseNot gradedqualityDmaintenanceMCP server for AWS S3 and compatible object stores, enabling LLM clients to browse buckets and fetch objects.MIT No Attribution