ltspice-mcp
Work in progress. Core functionality is usable but expect rough edges and breaking changes.
An MCP server that connects LLM assistants (Claude, and any other MCP client) to real circuit simulation: LTspice and ngspice, plus direct editing of LTspice .asc schematics. Simulation results come back as structured numbers — cutoff frequencies, overshoot, phase margin, rise times — so the assistant can design, verify, and iterate on circuits in the same files you open in LTspice. Built on spicelib.
Quick start
- Install the server:
uv tool install ltspice-mcp # or: pip install ltspice-mcp / pipx install ltspice-mcp
- Add it to your MCP client:
# Claude Code
claude mcp add -s project ltspice -- ltspice-mcp
// Claude Desktop (claude_desktop_config.json) and most other clients
{
"mcpServers": {
"ltspice": { "command": "ltspice-mcp", "args": [] }
}
}
- Have LTspice or ngspice installed. Both are auto-detected on Windows, Linux, and macOS; on WSL the LTspice path must be set explicitly (WSL notes). Circuit editing works with no simulator at all.
That's the setup. Python 3.11+ required. Verify with ltspice-mcp --help.
Claude Desktop config lives at ~/Library/Application Support/Claude/ (macOS), %APPDATA%\Claude\ (Windows), or ~/.config/Claude/ (Linux). Cursor, Windsurf, Gemini CLI, Continue, Cline, Zed and others take the same JSON snippet in their respective config files. Web clients (claude.ai, ChatGPT) need a stdio→HTTP bridge such as mcp-proxy — only expose this server on a network you fully control, since it writes files and spawns processes inside allowed_paths.
Install as a Claude Code plugin or Desktop Extension
Two lower-friction routes that wrap the same PyPI package:
- Claude Code plugin —
/plugin marketplace add cognitohazard/ltspice-mcp, then /plugin install ltspice-mcp. Registers the server (via uvx) and bundles the LTspice/ngspice skills. To test against a local checkout instead, /plugin marketplace add ..
- Claude Desktop extension — build the
.mcpb in packaging/mcpb/ and drag it onto Claude Desktop for a one-click install with a native folder picker for your circuits directory.
Both wrap the server; they still need uv and a simulator on the host — they do not bundle LTspice or ngspice.
Using it
Once connected, you ask for circuit work in plain language. The assistant chooses the tools; the server runs the simulator and measures the results.
"Design a 1 kHz RC low-pass filter and verify its cutoff."
The assistant writes the netlist, validates it, runs an AC simulation, and reads the measurements back: cutoff 1000.4 Hz, −19.9 dB/decade roll-off, first-order response. If the cutoff is off-target, it changes a component value and re-runs — each iteration is a couple of seconds.
Other requests that work the same way:
- "What's the overshoot and settling time of this regulator's step response?" — runs a transient analysis and measures both from the waveform, plus rise time, ringing frequency, and the final value.
- "Run a 200-run Monte Carlo with 5% resistors and tell me the output spread." — perturbs components per run, simulates the batch, and reports mean, sigma, and worst-case values per measurement.
- "Sweep the load from 100 Ω to 10 kΩ and find where efficiency drops." — parameter sweep with per-run results.
- "Build this differential pair as a schematic I can open in LTspice." — places and wires the components into a real
.asc, with orthogonal routing and pin-collision checks.
- "Is this loop stable?" — AC analysis of the loop gain; reports phase and gain margin at every crossover, not just the first.
Co-design on the same files
Everything operates on ordinary LTspice and SPICE files, so the work passes back and forth between you and the assistant instead of living inside a chat:
- Sketch a schematic in LTspice, then hand it over: "what's the bias point?", "why doesn't the output move?", "add compensation and check the phase margin."
- Or the reverse: the assistant designs and verifies the circuit and writes the
.asc; you open it in LTspice, inspect it, and tweak by hand. Your manual edits are simply the file's new state — the assistant picks up from there on the next request.
- Changes can flow either direction mid-design: adjust a value in the GUI and ask for re-verification, or have the assistant sweep a change you're considering before you commit to it.
What it does
