sbox-agent-bridge is a provider-neutral MCP server + s&box editor bridge. It gives agents a safe, observable way to operate the editor: read live scene state, make undoable changes, assign assets, run play mode, inspect compile/log feedback, and verify results from the editor instead of guessing from files.
Think: Claude/Codex/Kimi/etc. as a careful level-design and gameplay prototyping partner inside s&box — with narrow tools, read-backs, and smoke tests instead of an unsafe arbitrary-code bridge.
A normal coding agent can edit C# files, but it cannot see what actually happened in the s&box editor. This bridge closes that loop.
Quick start
Requirements: s&box installed, Node.js 20+, npm, and an MCP-capable client.
1. Build the MCP server
git clone https://github.com/user1303836/sbox-agent-bridge.git
cd sbox-agent-bridge\mcp-server
npm install
npm run build
2. Install the editor bridge into a project
For a fresh test project, create a Minimal Game project and install the bridge:
cd C:\path\to\sbox-agent-bridge
.\scripts\create-minimal-sbox-project.ps1 `
-ProjectPath 'C:\Users\you\Documents\s&box projects\AgentBridgeMvpFresh' `
-Title 'Agent Bridge MVP Fresh' `
-Ident 'agent_bridge_mvp_fresh'
.\scripts\install-editor-bridge.ps1 `
-ProjectPath 'C:\Users\you\Documents\s&box projects\AgentBridgeMvpFresh'
Or install into an existing s&box project:
.\scripts\install-editor-bridge.ps1 `
-ProjectPath 'C:\Users\you\Documents\s&box projects\YourProject'
Open the project in s&box and wait for compile. The bridge starts automatically after the editor library loads. A status dock appears under:
View -> Agent Bridge
3. Configure your MCP client
Point your client at the built server:
{
"mcpServers": {
"sbox-agent-bridge": {
"command": "node",
"args": ["C:/absolute/path/to/sbox-agent-bridge/mcp-server/dist/index.js"]
}
}
}
For isolated fresh-project or multi-editor runs, set the same IPC root for both the launched editor and MCP server:
{
"mcpServers": {
"sbox-agent-bridge": {
"command": "node",
"args": ["C:/absolute/path/to/sbox-agent-bridge/mcp-server/dist/index.js"],
"env": {
"SBOX_AGENT_BRIDGE_IPC": "C:/temp/sbox-agent-bridge-mvp-fresh"
}
}
}
}
Launch a project with that IPC root:
$env:SBOX_AGENT_BRIDGE_IPC = "$env:TEMP\sbox-agent-bridge-mvp-fresh"
.\scripts\start-sbox-project.ps1 `
-ProjectFile 'C:\Users\you\Documents\s&box projects\AgentBridgeMvpFresh\agent_bridge_mvp_fresh.sbproj' `
-IpcRoot $env:SBOX_AGENT_BRIDGE_IPC `
-ClearIpc `
-WaitForBridgeSeconds 90
4. Ask for the first read-only check
Use the sbox-agent-bridge MCP tools to run the bridge doctor, list editor tabs, summarize the active scene, and inspect the current selection. Do not mutate the scene yet.
Then try one tiny edit:
Create one GameObject named Agent Bridge Test, verify that it exists, then undo the creation.
What agents can do today
The current MCP tools are: editor, scene, gameobject, component, script, project, reference, asset, visual, sound, network, physics, prefab, and runtime.
For exact action payloads, see docs/protocol.md. For the verified capability matrix, see docs/capability-matrix.md.
A compelling workflow
Once the bridge is installed, an agent can perform a full editor loop:
- Run
bridge.doctorand read the active project/scene. - Create or recover a saved scene.
- Add GameObjects, components, models, materials, physics, sound, and prefabs.
- Validate component values before mutating them.
- Save the scene and enter play mode.
- Wait for runtime readiness and inspect the live
GameSession. - Run deterministic runtime test actions.
- Capture camera output and inspect compile/log feedback.
- Stop play mode and recover the editable scene.
That loop is what makes the bridge different from a file-only coding assistant: the agent can observe the editor, act, and verify.
Smoke tests
CI validates the TypeScript MCP server and metadata. Live editor behavior requires s&box to be open.
cd mcp-server
npm run ci # typecheck, unit tests, build
npm run smoke:mvp-suite # preferred external-tester live gate
npm run audit:capabilities
The MVP suite creates its own saved scene and verifies the main bridge path: doctor, compile wait, scene recovery, object creation, model/material assignment, asset type/resource/cloud reads, physics/sound/network/prefab read-back, runtime preview capture, scene metadata, spatial radius search, runtime component type discovery, undoable destruction, project file/input/reference helpers, script introspection and compile recovery, animation helper setup, particle setup, play/stop settle, and cleanup.
Other focused smokes:
npm run smoke:bootstrap
npm run smoke:matrix-core
npm run smoke:project
npm run smoke:reference
npm run smoke:scripts
npm run smoke:assets
npm run smoke:asset-resources
npm run smoke:physics
npm run smoke:prefabs
npm run smoke:sounds
npm run smoke:network
npm run smoke:capability-gaps
npm run smoke:runtime
npm run walkthrough:arpg
npm run walkthrough:boxing
npm run audit:capabilities
See docs/testing.md for all live-smoke commands and environment variables.
Current caveats
This is useful now, but still intentionally honest:
- The editor bridge must be installed into each s&box project that should expose live access.
- CI cannot currently run a real s&box editor, so editor behavior is verified by local smoke tests.
- Component discovery combines editor-visible TypeLibrary entries with runtime assembly scanning; exact-name
component.addremains the fallback for compiled local components. gameobject.duplicateis shallow: name/enabled/transform/parent, not full child/component cloning.visual.capture_cameracaptures camera output, not the full editor/game viewport overlay or generic UI panel hierarchy.- Semantic model orientation still needs stored overrides or human/vision confirmation; bounds alone cannot prove “upright.”
- There is no in-editor project switch action yet; use
scripts/start-sbox-project.ps1or open the.sbprojmanually.
See docs/status.md for current verification status and limitations.
Repository map
editor/ s&box editor library: dock, frame pump, dispatcher, handlers
mcp-server/ TypeScript MCP stdio server, bridge client, tools, tests
scripts/ project creation/install/launch helpers for tester workflows
schemas/ bridge command/response JSON schemas
docs/ status, protocol, testing, architecture, capability matrix
assets/ README/logo assets
Start with:
- Tester Quickstart — shortest validation path
- Docs Index — map of the detailed docs
- Status — verified state and caveats
- Capability Matrix — per-feature coverage
- Protocol — action names and payload notes
- Architecture — bridge design
Design principles
- Live state first: inspect the editor before mutating it.
- Small, named actions: no broad arbitrary-code bridge.
- Undoable when possible: scene edits use editor undo scopes.
- Verified read-back: mutations return what the editor actually sees after the change.
- Provider-neutral: works with any MCP-capable agent.
- s&box-grounded: no Unity/Godot/GMod API guessing.
License
MIT. See LICENSE.
