Arrgon
Turns the free Google Gemini Desktop App into an autonomous coding agent by exposing an MCP server over stdio or HTTP; Gemini calls file tools (list, read, write, patch, search, move, delete) and a sandboxed command runner to perform read โ edit โ verify coding loops on a local project workspace.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
๐ก What is Arrgon?
Arrgon is a local Model Context Protocol (MCP) server that gives Google Gemini the ability to read, write, search, and patch code files โ and run developer commands like flutter analyze, pytest, git diff, npm test, etc. โ all inside a sandboxed, zero-shell environment on your machine.
Use Gemini as a free AI coding agent. No API keys. No subscriptions. Just connect Arrgon to the Gemini Desktop App and start prompting.
What can you do with it?
๐ "Read
lib/main.dartand find all TODO comments in the project"โ๏ธ "Add error handling to the
fetchUser()function and runflutter analyzeto check for issues"๐งช "Write a unit test for the auth module, then run
pytestto verify it passes"๐ "Refactor
utils.pyto use dataclasses and checkgit diffto review the changes"
Gemini autonomously loops through read โ edit โ verify cycles until the task is done.
Related MCP server: MCPort
๐ Quickstart (2 Minutes)
Prerequisites
Python 3.10+ installed (python.org)
Gemini Desktop App installed (gemini.google.com/app)
Step 1: Clone & Install
git clone https://github.com/omkar-ukirde/Arrgon.git
cd Arrgon
python3 -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
pip install -r requirements.txtStep 2: Connect to Gemini Desktop App
Open the Gemini Desktop App โ Settings (gear icon) โ Connected Apps โ Add custom connected app.
You will see two setup methods. Use Local MCP (recommended):
Option A: Local MCP (Recommended)
The Gemini app launches and manages the server process for you. Zero maintenance.
Field | Value |
Setup method | Local MCP |
Server name |
|
Description |
|
Command | See below ๐ |
Sandbox | Full access |
Command โ paste the full path to start_mcp.sh:
/full/path/to/Arrgon/start_mcp.shExample (on macOS):
/Users/yourname/Arrgon/start_mcp.sh๐ก Changing your project folder:
Edit
DEFAULT_WORKSPACEinsidestart_mcp.shto point to whatever project directory you want to work on (e.g.,/Users/yourname/MyProject), orPass the directory directly as an argument in the Command field:
/Users/yourname/Arrgon/start_mcp.sh /Users/yourname/MyProjectAlways choose Full access under Sandbox settings so Arrgon can read and write to your workspace without permission errors.
Click Next โ Connect, and you're done!
Option B: Cloud MCP (HTTP Mode)
If you prefer running the server yourself in a terminal (useful for debugging or switching projects):
1. Start the server:
cd Arrgon
./run_server.sh /path/to/your/project2. In Gemini Desktop App, choose Cloud MCP and enter:
Field | Value |
URL |
|
Note: You must keep the terminal running. Closing it stops the server.
Step 3: Start Coding with Gemini
Go back to the Gemini chat and try:
"List all files in my project and give me an overview of the codebase structure"
Gemini will call Arrgon's tools to browse your files, read code, and respond with a structured overview. From there, you can ask it to make changes, run tests, debug errors โ anything a coding assistant would do.
โ๏ธ How It Works
โโโโโโโโโโโโโโโโโโโโโโโ MCP (stdio or HTTP) โโโโโโโโโโโโโโโโโโโโโโโโ
โ Gemini Desktop โ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโบ โ Arrgon โ
โ App (free) โ JSON-RPC tool calls โ MCP Server โ
โ โ โ โ
โ "Fix the bug in โ โโโบ read_file("src/app.py") โ Reads files โ
โ src/app.py and โ โโโ file content with line #s โ Writes atomically โ
โ run the tests" โ โโโบ patch_file(search, replace) โ Searches code โ
โ โ โโโ "Successfully patched" โ Runs safe commands โ
โ โ โโโบ run_command("pytest") โ โ
โ โ โโโ test results โ All sandboxed to โ
โ โ โ your workspace dir โ
โโโโโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโโโโโArrgon exposes 8 MCP tools that Gemini calls autonomously. Gemini decides which tools to call and in what order โ forming edit โ verify loops on its own.
๐ก๏ธ Security
Arrgon is designed to be safe for autonomous AI usage on your local machine. Every operation is sandboxed.
run_command โ Zero-Shell Terminal Runner
The run_command tool is the most security-critical component. Here's what makes it safe:
Layer | Protection |
| Commands are parsed with |
Binary whitelist | Only pre-approved developer tools are allowed: |
Subcommand whitelist | Even allowed binaries are restricted to safe subcommands. For example, |
Destructive token blocklist | Tokens like |
Hard timeout | Commands are killed after 60 seconds (configurable 5sโ180s) to prevent hangs from interactive prompts. |
Output truncation | Massive outputs are truncated to 15,000 characters so they don't crash Gemini's context window. |
Filesystem Sandbox
Layer | Protection |
Path traversal defense | All paths are resolved against |
Symlink escape detection | Paths are fully resolved with |
Atomic writes | Files are written to a temporary file first, then atomically swapped via |
Single-match patching |
|
๐ง Available Tools
Tool | Description |
| Lists workspace files with sizes, filtering out |
| Reads file content with numbered lines. Supports |
| Creates or replaces a file atomically. Auto-creates parent directories. |
| Exact search-and-replace on code. Requires a unique match to prevent unintended edits. |
| Regex or text search across all code files. Returns file paths, line numbers, and snippets. |
| Moves or renames files/directories within the workspace. |
| Deletes a single file (directory deletion is blocked for safety). |
| Sandboxed CLI runner for whitelisted dev commands (see Security). |
๐๏ธ Customization
Environment Variables
Variable | Default | Description |
|
| Root directory for all file operations |
|
| Host to bind HTTP server (Cloud MCP mode only) |
|
| Port for HTTP server (Cloud MCP mode only) |
Adding More Commands
Edit the ALLOWED_COMMANDS dictionary in mcp_coding_agent.py to allow more tools:
ALLOWED_COMMANDS: dict[str, Optional[set[str]]] = {
# Add your own:
"cargo": {"build", "test", "check", "clippy"},
"go": {"build", "test", "vet", "fmt"},
"make": {"test", "build", "lint"},
# ... existing entries ...
}Set a value of None to allow any subcommand for a binary (only for trusted tools like test runners):
"mypy": None, # Any arguments are allowedSwitching Projects
Local MCP: Edit the command in Gemini Desktop App settings to point to a different workspace directory.
Cloud MCP: Stop the server (Ctrl+C) and restart with a new path:
./run_server.sh /path/to/different/project๐งช Running Tests
Arrgon includes 18 automated tests covering command validation, path security, file operations, and HTTP endpoint integration:
source .venv/bin/activate
pytest -vtest_mcp_coding_agent.py::test_validate_command_allowed PASSED
test_mcp_coding_agent.py::test_validate_command_shell_operators_blocked PASSED
test_mcp_coding_agent.py::test_validate_command_blocked_tokens PASSED
test_mcp_coding_agent.py::test_validate_command_unauthorized_binaries_and_subcommands PASSED
test_mcp_coding_agent.py::test_validate_command_empty PASSED
test_mcp_coding_agent.py::test_resolve_safe_path_valid PASSED
test_mcp_coding_agent.py::test_resolve_safe_path_traversal_blocked PASSED
test_mcp_coding_agent.py::test_write_and_read_file PASSED
test_mcp_coding_agent.py::test_read_file_errors PASSED
test_mcp_coding_agent.py::test_patch_file PASSED
test_mcp_coding_agent.py::test_search_code PASSED
test_mcp_coding_agent.py::test_list_files PASSED
test_mcp_coding_agent.py::test_move_or_rename_file PASSED
test_mcp_coding_agent.py::test_delete_file PASSED
test_mcp_coding_agent.py::test_run_command_allowed_execution PASSED
test_mcp_coding_agent.py::test_run_command_blocked PASSED
test_mcp_coding_agent.py::test_run_command_invalid_directory PASSED
test_mcp_coding_agent.py::test_streamable_http_initialization_and_tool_list PASSED
============================== 18 passed ==============================๐ Project Structure
Arrgon/
โโโ mcp_coding_agent.py # Main MCP server (all tools + security)
โโโ test_mcp_coding_agent.py # Automated test suite (18 tests)
โโโ run_server.sh # Helper script for HTTP mode
โโโ requirements.txt # Python dependencies
โโโ LICENSE # MIT License
โโโ README.md # This file๐ค Contributing
Contributions are welcome! Some ideas:
Add more tool commands (Rust/Cargo, Go, Make, Docker, etc.)
Add a
create_directorytool for scaffolding projectsAdd
undo_patchwith file backup/restoreWindows support improvements (PATH handling, shell detection)
Multi-workspace support for monorepos
# Fork, clone, and run tests
git clone https://github.com/YOUR_USERNAME/Arrgon.git
cd Arrgon
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
pytest -v๐ License
MIT License โ use it however you want, free forever.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogโฆ
This server cannot be deployed
Maintenance
Related MCP Connectors
Persistent cloud workspaces for AI agents: run commands, edit files, use git and a browser.
Work on your own computers from Claude or ChatGPT: run commands, edit and search files.
Connect AI assistants to your approved computer for files, commands, browser and desktop tasks.
Securely search and manage workspace context files for AI agents and teams.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to develop within a local project workspace by reading and modifying files, running commands and tests, checking Git state, and persisting progress as history sessions that can be restored in later conversations.Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables AI clients to safely read, search, understand, and edit local project code and files, with Git inspection, code indexing, and controlled command execution within permissioned workspaces.8 npm6MIT
- AlicenseAqualityBmaintenanceEnables coding agents to safely inspect and modify project files within explicitly allowed workspace roots, with controls for file operations, git status/diff, and optionally allowlisted command execution.11Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables Gemini Spark to securely read, edit, and search a selected local workspace, execute shell commands, and integrate development tools via MCP with user approval safeguards.1MIT