> ## Documentation Index
> Fetch the complete documentation index at: https://docs.reactor.inc/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To build and serve your own model, start at /deploy/development/quickstart and /deploy/development/overview. Deploying is the default path: reactor init scaffolds a workspace, reactor auth login authenticates, and reactor model deploy registers the model, publishes the release with the weights/ folder, and activates it on Reactor's GPUs, in one command from that workspace. Docker must be running, because the publish step builds the image locally. Bump model.version in reactor.yaml before redeploying a change, because a release that already has an image is reactivated as it is. Deployment access is granted per account, so contact team@reactor.inc if a deploy is refused. Every key in reactor.yaml is documented at /deploy/platform/reactor-yaml. Model code imports reactor_runtime; Python client code imports reactor_sdk. The runtime overview explains the model interface. Running the model on your own machine with reactor run is optional and needs a GPU you attach with --gpus; /deploy/development/local-testing covers that loop and pairs a complete brightness model with a Python client test in a separate brightness-test workspace.
> Reactor hosts multiple models, each with its own connect slug (modelName) and command/event schema. The video model catalog — slug, typed SDK package, and links to its schema — is at /model-api-reference/overview. Robotics policy documentation starts at /robotics/overview; X-WAM observations, actions, and client integration are under /robotics/xwam/; Cosmos3 Nano Policy DROID is under /robotics/cosmos/nano-policy-droid/. Some models expose one slug per experience (e.g. HappyOyster); always take the slug from the model's own pages, never guess it.
> Fastest path to a working app: `npx create-reactor-app my-app --model=<slug>` scaffolds a complete app with secure auth wired up. Typed TypeScript SDKs are published as @reactor-models/<model>; Python uses the base reactor-sdk package.
> Auth: exchange an API key (rk_...) for a JWT via POST https://api.reactor.inc/tokens from your server. Never put the API key in client-side code.
> Append .md to any docs URL for clean Markdown. Search these docs via the MCP server at https://docs.reactor.inc/mcp.

# Troubleshooting

> Diagnose FastWAM connection, observation, and request failures.

| Symptom | Check |
| - | - |
| Model not found or no access | Confirm the model slug, hosted availability, and account access. |
| Session does not reach `READY` | Check SDK status/error messages, credentials, capacity, and endpoint. Do not send model commands before readiness. |
| State command acknowledged but no prediction | An acknowledgement is not an action reply. Set a nonblank task, inspect `command_error`, and keep both camera tracks publishing after the request. |
| `command_error` for `state_json` | Send a JSON string, eight finite numeric state values, and an integer `chunk_id`. Keep the serialized request within 2,000 characters. |
| Repeated request produces no reply | Identical JSON bytes are deduplicated. Increment the ID for a new observation; for a controlled retry, change a `retry` counter and deduplicate replies. |
| Actions look plausible but the task fails | Verify raw camera orientation, state coordinates, controller scaling, and gripper conversion. The service already rotates images and converts output actions. |
| Reply belongs to an old observation | Match request IDs, allow one request at a time, and discard timed-out results. Frame delivery is not synchronized capture. |
| No reply after reset | Send changed request bytes and fresh frames; resetting does not clear the task instruction or advance the robot. |

The helper unwraps message `data`. To inspect request rejections, subscribe to `command_error`
before connecting and consume that queue alongside predictions in your application. SDK command
errors and asynchronous model rejections are separate paths.

For a report, include SDK version, confirmed model slug, request ID, frame shapes, and the error
reason. Describe whether the observations were synthetic, simulated, or from hardware. Do not attach
API keys or private camera recordings.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.