Skip to main content
This guide walks you through everything you need to go from a brand-new account to a live, question-answering AI agent. By the end, you’ll have ingested your first piece of content, built a knowledge base around it, and deployed an agent you can share with others or embed on your site — all in about five minutes.
1

Sign Up and Create a Workspace

Before you build anything, you need a Beenos Solutions account and a workspace to work in.
  1. Go to app.beenossolutions.com and click Sign Up.
  2. Enter your email address and choose a password, or continue with Google or GitHub.
  3. After verifying your email, you’ll be prompted to create your first workspace. Give it a name that reflects your project or organisation — for example, Acme Support or Internal Docs.
Your workspace is the shared environment where all your knowledge bases, agents, and settings live. You can invite team members to it later.For a detailed walkthrough of account and workspace setup, see the Account Setup guide.
2

Add Your First Content Source

With your workspace ready, add the content you want your AI agent to know about.
  1. In the left sidebar, navigate to Knowledge Bases and click New Knowledge Base.
  2. Give your knowledge base a name, then click Add Source.
  3. Choose a source type:
    • Document — upload a PDF, Word file, or Markdown file from your computer.
    • URL — paste a public web page or sitemap and Beenos Solutions will crawl the content.
    • Plain Text — paste or type content directly into the editor.
  4. Click Process Source. Beenos Solutions will chunk, embed, and index your content automatically. Processing typically completes within a few seconds for small sources.
Start with a focused, well-structured document — like a product FAQ or a help article — to get the clearest results from your agent during testing.
3

Create an AI Agent and Connect It to Your Knowledge Base

Now create the agent that will query your knowledge base and respond to user questions.
  1. In the sidebar, go to Agents and click New Agent.
  2. Enter a name and an optional description for your agent.
  3. Under Knowledge Base, select the knowledge base you just created.
  4. Configure your agent’s behaviour:
    • Persona — give your agent a name and a short system prompt to shape its tone (e.g., “You are a friendly support assistant for Acme. Answer only from the provided knowledge base.”).
    • Response Style — choose between concise, detailed, or bullet-point responses.
    • Fallback Message — set the message your agent returns when it can’t find a relevant answer.
  5. Click Save Agent.
You can connect one agent to multiple knowledge bases, or create separate agents with different personas for different audiences — all from the same workspace.
4

Test Your Agent in the Chat Playground

Before sharing your agent, verify it’s responding correctly using the built-in playground.
  1. Open your agent and click the Playground tab.
  2. Type a question that your uploaded content should be able to answer.
  3. Review the response — check accuracy, tone, and whether the cited sources are correct.
  4. If a response is off, return to the Knowledge Base tab to review how the source was processed, or adjust the agent’s system prompt.
Responses in the Playground are live queries against your knowledge base and count toward your plan’s usage. Extensive testing on large knowledge bases may consume a significant portion of your monthly quota.
5

Embed or Share Your Agent

Once you’re happy with how your agent responds, deploy it for your users.Choose one of the following deployment options from the Deploy tab on your agent’s page:
  • Hosted Chat Link — share a public URL that opens a full-page chat interface. No code required.
  • Embed Widget — copy and paste a single <script> tag into your website’s HTML to add a chat bubble.
  • REST API — send queries programmatically and integrate agent responses into your own application or workflow.
Your agent goes live immediately. Any updates you make to the connected knowledge base are reflected in agent responses automatically.
If you want to manage knowledge bases, ingest content, or query agents programmatically, head to the API Reference for full endpoint documentation, request examples, and authentication details.