Skip to main content
Glama

Local Model Workers MCP

Local Model Workers MCP is a local MCP server under development that will let Claude Code and Codex delegate repository exploration and test proposals to a model served by LM Studio on another machine in a private local network.

The local server remains the security boundary: it selects repository context, enforces path and content restrictions, validates remote output, and returns a structured result. The remote model never writes to the repository, applies a patch, or runs a project command.

Project status

The product requirements and implementation plan are approved. The project foundation is implemented: the repository has a pinned TypeScript/Node.js toolchain, a buildable CLI entry point, automated tests, and a CI validation workflow. Product tools, the MCP server, configuration files, and the environment-variable contract are not implemented yet.

The intended V1 exposes exactly six MCP tools:

  • explore_repository

  • propose_tests

  • check_health

  • get_config

  • validate_config

  • update_config

See prd.md for the complete requirements and acceptance criteria.

Related MCP server: mcp-flow

V1 boundaries

  • The MCP server runs locally over stdio.

  • Only inference traffic reaches LM Studio over authenticated HTTP on a trusted private network.

  • Repository access is read-only and restricted to the requested root.

  • Sensitive, ignored, binary, and explicitly excluded files are never sent to LM Studio.

  • Test generation returns a validated unified diff; the MCP server never applies it or executes tests.

  • Task content is ephemeral. Operational logs contain metadata only and are retained for seven days.

  • macOS is the fully validated platform for V1. Basic portability checks are planned for Linux and Windows.

Documentation

Development

Use Node.js 24.18.0 and npm 11.16.0. With nvm, the repository baseline can be selected from .nvmrc.

nvm use
npm ci
npm run validate

npm run validate checks formatting, linting, static types, tests, and the production build. The current minimal CLI can also be built and inspected:

npm run build
node dist/cli/index.js --version

The version is written to stderr so stdout remains reserved for the future MCP stdio protocol. Running the CLI without arguments currently exits cleanly and does not start a server.

Do not create a local .env from guessed variable names. The supported names, validation rules, and a redacted .env.example must be added together when the configuration contract is implemented.

F
license - not found
-
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • A
    license
    A
    quality
    C
    maintenance
    A local MCP server that lets Claude delegate scoped work to Codex with structured results and guardrails, supporting planning, code review, build, reverse engineering, and long-running background tasks.
    Last updated
    11
    MIT
  • F
    license
    -
    quality
    C
    maintenance
    A safe, local MCP server that lets Claude drive a controlled software-development loop (inspect, read, plan, patch, apply, check, analyze, fix, summarize) on a project, using deterministic tools and real diffs/test runs.
    Last updated
    1
  • A
    license
    A
    quality
    B
    maintenance
    A read-only MCP server that lets a Claude chat explore your local repository and answer questions about it, returning synthesized answers with file:line references.
    Last updated
    3
    1
    MIT

View all related MCP servers

Related MCP Connectors

  • Augments MCP Server - A comprehensive framework documentation provider for Claude Code

  • An MCP server that gives your AI access to the source code and docs of all public github repos

  • Hosted MCP server connecting claude.ai, ChatGPT and other AI apps to your own computer

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/gaabrielrd/local-model-workers-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server