Quick Start

Sign up, save a provider key, ship an agent, and send your first API request — all in under ten minutes.

The 10-minute path

Onboarding checklist · app.sentientone.ai
Welcome to SentientOneStep 4 of 5Continue
Your account

Email

admin@acme.com

Plan

Pro trial · 14 days remaining

Platform API key

sk-so-•••• 3F92
1 agent1 saved key0 requests

What you'll have at the end

A live agent on the SentientOne platform, a saved LLM provider key, your platform API key, and a working terminal command that returns an AI response from your own configured agent.

Step-by-step

  1. 1

    Create an account

    Sign up at app.sentientone.ai. You land on the dashboard with a 14-day trial of Pro already active — no credit card required.
  2. 2

    Save your first LLM provider key

    Open LLM Keys in the sidebar. Click + Add key, pick a provider (OpenAI, Anthropic, Gemini, or Groq), name the key, paste it in, and save. Keys are encrypted at rest and only used when an agent of yours calls that provider.
  3. 3

    Create an agent

    Click Create Agent in the global header to open the create wizard. Pick a template (Website Chatbot is a safe first pick), tweak the name, choose your provider, then pick a model from the Model dropdown of that provider's current models. Select the saved key from the Use saved key dropdown and finish the wizard — you land on the new agent's Agent Settings view.
  4. 4

    Copy the Agent ID

    On the Agent Settings view, the page header carries a click-to-copy Agent Id. Click to copy it — you'll send it in every API request as the X-Agent-Id header.
  5. 5

    Copy your platform API key

    Open the user menu (your avatar, bottom-left of the sidebar) and choose SettingsAPI Key. Reveal the key (starts with sk-so-) and copy it. Every request to the platform sends this as the X-Api-Key header.
  6. 6

    Send your first request

    From any terminal, run the curl below. Replace the two placeholders with the values you just copied.
    bash
    curl -N -X POST https://api.sentientone.ai/v1/chat/stream \
      -H "Content-Type: application/json" \
      -H "X-Api-Key: sk-so-YOUR_KEY" \
      -H "X-Agent-Id: YOUR_AGENT_ID" \
      -d '{ "message": "Hello!" }'
  7. 7

    See the response

    The endpoint streams Server-Sent Events — one JSON object per data: line. You'll see a meta event with the conversation_id, an optional sources event, a run of delta events, and a final done event. The full answer is the concatenation of every delta event's content.
    text
    data: {"type":"meta","conversation_id":"...","trace_id":"..."}
    data: {"type":"sources","sources":[{"index":1,"id":"...","title":"...","url":"...","source_type":"...","score":0.91,"snippet":"..."}]}
    data: {"type":"delta","content":"Hel"}
    data: {"type":"delta","content":"lo!"}
    data: {"type":"done","conversation_id":"...","trace_id":"..."}
    Keep the conversation_id and pass it back on the next request to continue the conversation.

Stay on the dashboard and learn how each surface works — Agents, Playground, ChatBot. Or jump straight into the REST API if you're wiring an integration.

Tip

Use the in-app API Sandbox to verify a request before pasting it into your codebase. It hits the same endpoint as production and shows you the full response, status, headers, and timing.