Get started
Setup prompts
Use these instructions to configure an application or coding client without putting a secret in the prompt.
Connect an application
Integration prompt
Connect this application to an OpenAI-compatible Chat Completions API.
Base URL: https://uncensoredlabs.dev/v1
API key: read UNCENSOREDLABS_API_KEY from the server environment
Model: choose an enabled ID from GET /v1/models
Use the existing OpenAI SDK when available. Keep the key on the server,
never in browser code, logs, or source control. Set max_tokens to 256
for the first check. Support SSE when stream is true. Handle structured
errors and uncertain responses without blindly retrying a billable call.
Show the configuration changes before making a test request. Do not
change unrelated providers or assume Responses, images, or embeddings
are supported.Configure a coding client
Client setup prompt
Configure this client's OpenAI-compatible Chat Completions provider.
Use https://uncensoredlabs.dev/v1 and the UNCENSOREDLABS_API_KEY environment
variable. Ask me for the approved model ID if it is not configured.
Confirm that the client's selected mode uses Chat Completions. Check the
model's streaming and tool-calling capabilities before enabling them.
Preserve the existing configuration so I can switch back. Never print
or commit the key. Report any client feature that needs another API.Verify the result
- Replace the example API domain with your own deployment's domain.
- Review the client's chosen provider mode and model ID.
- Make one small request, then inspect its status, tokens, and charge in Usage / Activity.