Atlas / MCP servers / luckysolanki902 / local-terminal-mcp

local-terminal-mcpSAFE

Name: local-terminal-mcp
Author: luckysolanki902

mcp/luckysolanki902/local-terminal-mcp

MCP server by luckysolanki902 · ai tools · automation · chatgpt · Audited 2026-10-11

Run local commands from ChatGPT or Claude via a safe, allowlisted MCP server. Work on your codebase and generate images/sprites/assets locally from ChatGPT without spending Codex/API tokens. stdio for Claude Desktop/Code; capability-URL or bearer auth over HTTP for ChatGPT via ngrok or cloudflared.

Verdict

SAFE

Grade

Trust score

89 /100

Exposed tools

3 3r · 0w · 0d

Transport

stdio · streamable-http

License

MIT

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

A small, security-first MCP server that lets an AI assistant — ChatGPT, Claude Desktop, Claude Code, or any other MCP client — run a set of allowlisted commands and read files from a local directory on your machine.

The motivating use case: point ChatGPT (or Claude) at a local codebase so it can run git, rg, cat, etc. and analyse the repo — without copying files around by hand, and, for ChatGPT, using your flat-rate subscription instead of metered API/agent tokens.

⚠️ This runs commands on your computer. It is built to be safe by default (read-only, allowlisted, no shell, path-contained, auth-required over the network), but you are responsible for how you configure and expose it. Read Security model before exposing it to the internet.

🚀 Just want to get going? Follow the install & setup checklist.

Why this exists
How it works
Security model
Install
Quick start (local / stdio)
Connect to Claude Desktop
Connect to Claude Code
Connect to ChatGPT (over the internet) · exact step-by-step guide
Configuration reference
Tools exposed
Development
FAQ

Why this exists

Running a coding agent against a large repo through a metered API burns tokens fast. If you already pay for a ChatGPT or Claude subscription, you can instead give the assistant a tool to read files and run read-only commands on your machine, and let it analyse the code conversationally. That is what this server provides, with a security model strong enough that you can leave write access turned off and expose only read access.

What

Read from source at commit effebf30c383OBSERVED · 2026-10-11

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code

claude mcp add local-terminal-mcp -- uvx local-terminal-mcp

claude-desktop

{
  "mcpServers": {
    "local-terminal-mcp": {
      "command": "uvx",
      "args": [
        "local-terminal-mcp"
      ]
    }
  }
}

Exposed tools (3)

3 read · 0 write · 0 destructive.

Tool	Risk	Description
`close_terminal`	read	return (
`list_directory`	read	try:
`list_terminals`	read	lines = [

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: none-observed
Network: declared (3 observation(s))
Shell: none-observed
Dependencies: pinned
Secrets in source: none-found

Findings (12)

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

tests/test_matching.py:84

["cat", "../../etc/passwd"],

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

tests/test_matching.py:85

["cat", "src/../../outside"],

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

tests/test_policy.py:126

policy.resolve_path("../../etc/passwd")