LMStudio BridgeSAFE
A Model Control Protocol (MCP) server that allows Claude to communicate with locally running LLM models via LM Studio.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
A Model Control Protocol (MCP) server that allows Claude to communicate with locally running LLM models via LM Studio.
Overview
LMStudio-MCP creates a bridge between Claude (with MCP capabilities) and your locally running LM Studio instance. This allows Claude to:
- Check the health of your LM Studio API
- List available models
- Get the currently loaded model
- Generate chat and raw text completions using your local models
- Generate vector embeddings for semantic search and RAG
- Hold stateful multi-turn conversations via response IDs
- Start and continue persistent conversations with a locked-in system prompt
This enables you to leverage your own locally running models through Claude's interface, combining Claude's capabilities with your private models.
Prerequisites
- Python 3.7+
- LM Studio installed and running locally with a model loaded
- Claude with MCP access
- Required Python packages (see Installation)
🚀 Quick Installation
One-Line Install (Recommended)
curl -fsSL https://raw.githubusercontent.com/infinitimeless/LMStudio-MCP/main/install.sh | bash
Manual Installation Methods
1. Local Python Installation
git clone https://github.com/infinitimeless/LMStudio-MCP.git cd LMStudio-MCP pip install requests "mcp[cli]" openai
2. Docker Installation
# Using pre-built image docker run -it --network host ghcr.io/infinitimeless/lmstudio-mcp:latest # Or build locally git clone https://github.com/infinitimeless/LMStudio-MCP.git cd LMStudio-MCP docker build -t lmstudio-mcp . docker run -it --network host lmstudio-mcp
3. Docker Compose
git clone https://github.com/infinitimeless/LMStudio-MCP.git cd LMStudio-MCP docker-compose up -d
For detailed deployment instructions, see [DOCKER.md](DOCKER.md
24f75fb5cbf4OBSERVED · 2026-10-06Exposed tools (9)
7 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
chat_completion | read | Generate a chat completion from the current LM Studio model. |
continue_conversation | read | Continue a stateful conversation started with start_conversation. |
create_response | write | Create a stateful response using LM Studio |
generate_embeddings | read | Generate vector embeddings for text using LM Studio. |
get_current_model | read | Get the currently loaded model in LM Studio. |
health_check | read | Check if LM Studio API is accessible. |
list_models | read | List all available models in LM Studio. |
start_conversation | write | Start a stateful conversation with a persistent system prompt. |
text_completion | read | Generate a raw text completion (non-chat format) from LM Studio. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (10 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (6)
LMSTUDIO_API_BASE=http://127.0.0.1:1234/v1 python lmstudio_bridge.py
-e LMSTUDIO_API_BASE=http://127.0.0.1:1234/v1 \
- In `lmstudio_bridge.py`, change `http://localhost:1234/v1` to `http://127.0.0.1:1234/v1`
requests, mcp, openai
curl -fsSL https://raw.githubusercontent.com/infinitimeless/LMStudio-MCP/main/install.sh | bash
curl -fsSL https://raw.githubusercontent.com/infinitimeless/LMStudio-MCP/main/install.sh | bash
Gates applied: no_behavioural_pass.
24f75fb5cbf4full audit observations/trust-audit/mcp-server/infinitimeless__lmstudio-bridge.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | 24f75fb5cbf4 | SAFE | B | 89 | first audit |
Questions
What is the LMStudio Bridge MCP server?
A Model Control Protocol (MCP) server that allows Claude to communicate with locally running LLM models via LM Studio.
What tools does LMStudio Bridge expose?
9 in total: 7 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is LMStudio Bridge safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does LMStudio Bridge need?
No credential environment variables were found in its source, so it appears to need none.
How does LMStudio Bridge run?
It speaks stdio, so it runs as a local process your client starts.
How current is this page?
The grade is for one exact copy of the source (24f75fb5cbf4), read on 2026-10-06. The repository is watched and re-audited when it changes.