# Occam

Occam finds mathematical equations from numeric records. It helps recover differential equations for dynamic systems or fit closed-form algebraic expressions from data.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 195 installs
- Author: occam
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/occam/occam
- Updated: 2026-04-05
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/occam

## What it does

Occam accepts sets of numerical observations and discovers mathematical relationships behind them. The `sindy.run` tool extracts governing differential equations from multidimensional time series. It finishes in seconds and outputs readable sparse formulas for dynamical models.

For static feature-target pairs, the server provides `pysr.run`. It uses evolutionary symbolic regression to find closed-form algebraic equations. Because an evaluation takes between 10 and 60 seconds, results take longer than SINDy. The response organizes formulas along a Pareto front balancing complexity against precision.

Users can suggest missing capabilities via `feature.request`. Calculations run remotely over HTTP through `https://occam--occam.run.tools`.

## Requirements
- Remote server: `https://occam--occam.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http occam https://occam--occam.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add occam --url https://occam--occam.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "occam": {
            "url": "https://occam--occam.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "occam": {
            "type": "http",
            "url": "https://occam--occam.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http occam https://occam--occam.run.tools
```

## When to use it
- extract differential equations from time series records
- fit closed-form algebraic equations with symbolic regression
- evaluate equation trade-offs along a Pareto frontier
- log feature requests for unsupported capabilities

## Example prompts
- Discover differential equations for this time series using sindy.run
- Find an algebraic relationship between features and targets via pysr.run
- Submit a feature request through feature.request for custom loss functions

## Similar
- [symbolic-equation](https://affpapa.org/en/nevibecode/skills/symbolic-equation.md) — skill · 1600 installs
- [arxiv-search](https://affpapa.org/en/nevibecode/skills/arxiv-search.md) — skill · 1100 installs
- [math-tools](https://affpapa.org/en/nevibecode/skills/math-tools.md) — skill · 1 install
- [fields](https://affpapa.org/en/nevibecode/skills/fields.md) — skill · ★ 3951 repo stars
- [sigma-algebras](https://affpapa.org/en/nevibecode/skills/sigma-algebras.md) — skill · ★ 3951 repo stars
- [maxima-cas-interface](https://affpapa.org/en/nevibecode/skills/maxima-cas-interface.md) — skill · 1 install

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
