How to Embed or Integrate Promptly Chatbots and Apps Into Your Website

Promptly gives you three integration paths: an embeddable widget you paste into your site, APIs that run your app from your own code, and pre-built connectors for Slack and Discord. The widget route requires no coding and is the fastest; the API route is for teams that want AI capabilities inside an existing product. All three are described on the Promptly site as part of the same no-code platform, so you can start with the widget and move to APIs later without rebuilding your app.

Option 1: Embed the widget with a code snippet

This is the path Promptly describes for adding a chatbot to a website. The steps map to the platform's own "How it works" flow:

1. Build the app first. Pick a template for your use case, provide the input and configuration it asks for, and choose how the output is rendered.
2. Attach your data. Add sources such as files, website URLs, sitemaps, YouTube links, Google Drive, or Notion exports, then attach them to the app in the app builder.
3. Publish. Save and publish the app. Promptly states you can then access it from a dedicated app page or embed it in your website using the embed code.
4. Paste the snippet. Copy and paste the provided code snippet where you want the chatbot to appear.
5. Match your branding. Promptly says you can customize the chatbot's look and feel to match your website, including a logo for the bot.

Expected result: a conversational AI widget running on your site, backed by the data sources you attached, without writing a RAG pipeline yourself.

Common snag: the widget only knows what you attached. If answers look thin or off-topic, revisit step 2 before touching the embed code.

Option 2: Run the app through Promptly's APIs

If you want AI inside an existing application rather than a floating widget, Promptly provides APIs for running apps and endpoints, plus the ability to upload and manage data programmatically. The site frames this as integrating AI capabilities into applications you already have — for example adding natural language processing to a chatbot, automating content moderation, or building intelligent recommendations into a product.

Use this route when:

• The AI output needs to appear inside your own UI, not in a Promptly-hosted widget.
• You need to trigger the app from a backend service or workflow.
• You want to manage data sources through code instead of the app builder.

Option 3: Slack apps and Discord bots

Promptly states you can build GPT-based Slack apps and Discord bots backed by your data without coding RAG pipelines. This is the right choice when your users already live in a chat tool and you don't want to drive them to a separate web page. The data-attachment step is the same as the widget path — your sources feed the bot's answers.

Which integration should you pick?

What to check before you start

• Model choice. Promptly supports major model providers including OpenAI, Cohere, Stability AI, and Hugging Face, and its site notes Google PALM models are available. Model chaining is supported, so a single app can route through more than one provider.
• Data readiness. The quality of your embedded chatbot depends on the sources you attach. Web URLs, sitemaps, PDFs, audio, PPTs, Google Drive, Notion imports, and YouTube URLs are all listed as supported inputs.
• Pricing and access. The site references a pricing page and an enterprise plan, and offers "Get Started for Free" and "Book a Demo." Check the pricing page directly for current plan limits before assuming a given integration path is included at your tier.

The practical sequence most teams follow: build and test the app in the builder, verify answers against your attached data, publish, then choose widget, API, or chat-platform delivery based on where your users already are.