M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
K

kubefwd

Community
Bulk Kubernetes service forwarding for local development.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #394 of 438
Stars
★ 4.2k Very popular
Transport
stdio (local process)
Runtime
Prebuilt binary · Docker
Credentials
No credential needed
License
Apache-2.0
Last commit
34FMRS · D

Based on the repository, server manifest, and README, kubefwd is described as a Kubernetes bulk port-forwarding tool for local development. The manifest provides stdio MCP packages, while the README mentions Claude Code and Cursor integration and an Apache-2.0 license. No specific MCP tool identifiers or client configuration examples are provided.

Strongest · Documentation 10/20 Weakest · Reliability 2/20

Reliability
2/20
Security and permissions
6/20
Maintenance
9/20
Documentation
10/20
Setup experience
7/20
Why each score
Reliability 2/20
The materials provide only stdio MCPB package entries, with no declared tools, initialization/tool-list handshake evidence, tests, or CI evidence. The version is 0.0.0, package hashes remain PLACEHOLDER_SHA256, and the declared capabilities cannot be checked against real behavior. The score is therefore very low, with deductions for unverifiable installation, error handling, and key paths.
Security and permissions 6/20
The documentation indicates Kubernetes cluster access, sudo/root privileges, modification of /etc/hosts and network interfaces, and local port forwarding. It also mentions a REST API, log streaming, and AI integration. However, there is no evidence of permission scoping, credential handling, data-flow disclosure, API binding restrictions, or confirmation for risky operations. No explicit listed red-line behavior is established, so this is not scored in the 0-4 range, but substantial points are deducted for high privilege and side effects without documented controls.
Maintenance 9/20
The repository is not archived, the license is explicitly Apache-2.0, and the README identifies a maintainer, sponsor, contribution guidance, and OpenSSF/SLSA-related badges, providing some maintenance and security-practice signals. However, the supplied source contains no verifiable release history, commit activity, dependency update process, issue-response evidence, or formal security-response channel; the manifest version is still 0.0.0. This supports only a modest score.
Documentation 10/20
The README covers purpose, operation, features, platform installation, examples, prerequisites, and links to REST API and MCP integration documentation. The supplied material does not show actual MCP tool names, parameters, return values, error behavior, permission boundaries, cost/resource limits, or client configuration. The manifest package information is also unverified, so points are deducted for incomplete operational evidence.
Setup experience 7/20
The README gives installation paths for macOS, Linux, Windows, and Docker, plus basic CLI examples, and clearly states the kubectl and root/sudo prerequisites. However, the manifest lists only macOS and Windows MCPB packages, hashes are placeholders, and no mainstream MCP client connection configuration or verifiable end-to-end setup is provided. Root access and manual environment preparation further increase setup complexity.

Static review · not runListed 2026-07-29

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers who need local access to Kubernetes services.
  • Teams seeking to reduce environment-specific connection configuration.
  • Users who need to forward multiple services or avoid local port conflicts.

Not for

  • Local development without a Kubernetes cluster.
  • Environments without cluster access or the ability to grant root/sudo privileges.
  • Simple single-service forwarding where service-name resolution, auto-reconnect, and monitoring are unnecessary.

Required permissions

  • A kubectl configuration with access to the target Kubernetes cluster is required.
  • Root/sudo access is required to modify /etc/hosts and network interfaces.
  • Access through the Kubernetes API is required to establish service port forwards.

Risks and side effects

  • The tool modifies the local /etc/hosts file and network interfaces, which can affect local service-name resolution.
  • Cluster services are forwarded to the local machine, so local applications may access services in the development cluster.
  • The tool requires privileged execution; its permissions should be appropriate for the target development environment.

Setup

Before you start

Runtime：Prebuilt binary · Docker

On macOS, run brew install kubefwd. On Linux, download a deb, rpm, or tar.gz package from releases. On Windows, use winget install txn2.kubefwd or Scoop. Docker is also supported. kubectl must be configured with access to a Kubernetes cluster before use.

Check that it works

After installing, confirm the MCP integration is available per kubefwd.com/mcp-integration/, then run sudo -E kubefwd svc -n default --tui and check that services receive unique 127.x.x.x IPs and curl http://api-service:8080 resolves by service name.

Troubleshooting

  1. Verify that kubectl is configured and can access the target cluster.
  2. Verify that the runtime environment has root/sudo access.
  3. Check that /etc/hosts and network interfaces are writable.
  4. Confirm the target namespace, label selectors, and service ports.
  5. If connections drop, check pod status and cluster connectivity; kubefwd supports automatic reconnection.

Things to try

Once connected, you can ask your AI assistant things like:

  • Bulk port-forward all services in the default namespace to my local machine.
  • Forward services from both the default and staging namespaces.
  • Only forward services labeled app=api.
  • Show the currently forwarded services and their traffic metrics.

Use cases

Run an application locally while accessing databases, authentication services, caches, or APIs in a development cluster.
Bulk-forward services in a namespace and access them by service name.
Use the TUI to inspect service status, traffic metrics, and container logs.
Control forwarding programmatically through the REST API.

Supported clients

Claude CodePartial support
CursorPartial support

Listed from the project's documentation, not tested by this site.

Overview

kubefwd is a command-line utility that bulk-forwards Kubernetes services to a local workstation. It assigns each service a unique 127.x.x.x loopback IP and automatically adds service names to /etc/hosts. It includes an interactive TUI, automatic reconnection, traffic monitoring, pod log streaming, a REST API, and AI assistant integration.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 7ea0970db0df Data synced 2026-10-19 Read the FMRS scoring method