Stochastly Desktop

Official

by MaheChadeffaud · Python

Local MCP access to Stochastly desktop nodes, diagnostics and research tools.

Model Context Protocol (MCP) Server: com.stochastly/desktop

The com.stochastly/desktop MCP server provides local MCP access to Stochastly desktop nodes, including diagnostics and research tools. It is focused on connecting a local environment to Stochastly desktop components so clients can use the associated node capabilities through MCP.

🛠️ Key Features

Local MCP access
Stochastly desktop nodes
Diagnostics tools
Research tools

🚀 Use Cases

Running diagnostics against Stochastly desktop nodes
Using research tooling tied to Stochastly desktop components via MCP

⚡ Developer Benefits

Clear separation between local Stochastly desktop resources and MCP clients
Centralized access to diagnostics and research tooling for desktop nodes

⚠️ Limitations

Scope is limited to local MCP access for Stochastly desktop nodes; no additional environments are specified

Topics

backtestingmcpmcp-servermodel-context-protocolquanttrading

README.md2 KBView on GitHub

Stochastly MCP

This small public launcher connects an MCP client to the server shipped with the Stochastly desktop app. The app must be installed, running, and licensed for Atelier or Portfolio. The package contains no Stochastly engine code.

Install with pip install stochastly-mcp or run with uvx stochastly-mcp. To try the unpublished source, install this directory with pip install ..

Configure a local stdio server with command stochastly-mcp and no arguments:

Claude Desktop: add "stochastly": {"command": "stochastly-mcp"} under mcpServers in claude_desktop_config.json.
Cursor: add the same entry under mcpServers in .cursor/mcp.json.
VS Code: add "stochastly": {"type": "stdio", "command": "stochastly-mcp"} under servers in .vscode/mcp.json.

The server exposes 29 tools, among them node_catalog, compose_graph, backtest_graph, red_flag_check, adversarial_check, portfolio_combine and export_paper. The installed app runs them and checks the licence on every call: portfolio_combine needs Portfolio, every other tool Atelier or Portfolio.

Without the app, the launcher lists the same 29 tools (names, descriptions, input schemas and annotations from stochastly_mcp_data/tools_manifest.json, generated from the app's server), so a client can see what the server offers. Each description then starts with a note that the desktop app is required, and every call returns a structured answer with ok: false, the edition needed and a link to pricing. The manifest is metadata only.

For a nonstandard installation, set STOCHASTLY_MCP_SCRIPT to the installed app's stochastly_mcp.py. This is a local path, not a network endpoint.

Install

Configuration

Environment variables

STOCHASTLY_MCP_SCRIPT

Optional path to the MCP client script supplied by the installed Stochastly app.

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "desktop": {
      "command": "uvx",
      "args": [
        "stochastly-mcp"
      ],
      "env": {
        "STOCHASTLY_MCP_SCRIPT": "<YOUR_STOCHASTLY_MCP_SCRIPT>"
      }
    }
  }
}