---
title: "Agent getting started: set up Spatial Compiler in Claude Code, Codex or Cursor"
description: "Set up Spatial Compiler in a coding agent: install the skill, connect the hosted MCP server, then dry-run, check and bake with pyro and read results as JSON."
canonical: https://appliedmanifold.com/docs/agents/
site: "Applied Manifold: The spatial compute cloud for 3D reality"
---

1.  [Home](https://appliedmanifold.com/)
2.  [Docs](https://appliedmanifold.com/docs/)
3.  Agent getting started

DOCS // SPATIAL.COMPILER

# Get started with an agent

Give your coding agent the Spatial Compiler skill, connect it to the hosted MCP server, then let it dry-run, check and bake. The same steps work in Claude Code, Codex, Cursor and other agents.

## 1\. Install the skill

Paste this prompt into your agent. It installs the Spatial Compiler skill and links its SKILL.md.

**Give this to your agent** (Claude Code, Codex, Cursor, others)

```
$ set up Spatial Compiler - https://appliedmanifold.com/llms.txt
```

Agent prompt:

Install the Spatial Compiler skill. If you're in Claude Code, run `claude plugin marketplace add applied-manifold/skills`, then `claude plugin install spatial-compiler@applied-manifold`. If you're in another agent, run `npx skills add applied-manifold/skills --skill spatial-compiler`. You can read the skill directly at https://github.com/applied-manifold/skills/blob/main/skills/spatial-compiler/SKILL.md (raw: https://raw.githubusercontent.com/applied-manifold/skills/main/skills/spatial-compiler/SKILL.md). Then use the Spatial Compiler skill when working on this project.

## 2\. Connect over MCP

The hosted server signs you in with an OAuth 2.1 browser login, so the config holds no key:

mcp.json

```json
{
  "mcpServers": {
    "spatial-compiler": {
      "type": "http",
      "url": "https://mcp.spatialcompiler.dev"
    }
  }
}
```

In Claude Code you can register it from the command line:

terminal

```sh
claude mcp add --transport http spatial-compiler https://mcp.spatialcompiler.dev
```

For the local server or REST, use an API key: see [Install via MCP](https://appliedmanifold.com/docs/mcp/) and [API keys](https://appliedmanifold.com/docs/api-keys/).

## 3\. The loop

1.  Pre-flight with `dry_run: true`. It costs $0 and returns `cost_usd_est`.
2.  For a pyro effect, copy the nearest of the [pyro examples](https://appliedmanifold.com/docs/pyro/examples/) and run `spatial-compiler check` on it. Each error names the key and the nearest valid key or the range.
3.  Bake at draft quality, read `warnings` in the report and change the key each `fix.hint` names.
4.  Open `preview_strip.png` and say what you see, since only the image shows whether the effect reads as fire.
5.  Bake at final quality until the report shows `"ok": true` and an empty `warnings` list.

## 4\. Read the docs as text

Every docs page has a markdown twin (this one: [/docs/agents/index.md](https://appliedmanifold.com/docs/agents/index.md)), also served for `Accept: text/markdown`. [/llms.txt](https://appliedmanifold.com/llms.txt) lists every page with its one-line answer, and [/llms-full.txt](https://appliedmanifold.com/llms-full.txt) holds the full text. See [Machine-readable docs](https://appliedmanifold.com/docs/machine-readable/).

## Agent offer

Agent offer: the first 200 Pro or Studio subscriptions get 50% off the first month with code AGENT-DOCS. Ends January 31, 2027.
