thotis.ai
Sign Up

Getting Started

Set up your first AI agent on Thotis RAG in minutes.

Prerequisites

  • A Thotis account with a team workspace
  • At least one dataset uploaded and indexed in RagFlow

Step 1 — Access the RAG console

Navigate to your team workspace and select RAG in the sidebar. This is the control plane where you manage all your agents and datasets.

Step 2 — Create your first agent

  1. Go to Agents and click New Agent.
  2. Fill in:
    • Name: a human-readable label (e.g. "Orientation Parcoursup")
    • Key: a URL-safe slug (auto-generated from the name)
    • Engine type: choose chat for a simple Q&A assistant or agent for a workflow-based agent
  3. Write a system prompt that defines the agent's behavior and tone.
  4. Optionally set an opener message shown when a new conversation starts.

Step 3 — Bind datasets

Go to the agent's Sources tab and attach one or more datasets. These are the knowledge bases the agent will search when answering questions.

Step 4 — Test in the Playground

Open the Playground from the sidebar. Select your agent and send a test message. Verify that the agent retrieves relevant chunks and produces a useful answer.

Step 5 — Publish

When you are satisfied with the agent's behavior:

  1. Go to the agent detail page.
  2. Click Publish. This pushes the configuration to RagFlow and creates a versioned runtime manifest.
  3. If the mobile backend is configured, a sync webhook delivers the manifest automatically.

Next steps

  • Fine-tune the agent's LLM settings in the Settings tab (model, temperature, top_p).
  • Set up the embed widget to expose the agent on your website.
  • Monitor performance in the Analytics dashboard.