Skip to main content

Look below

The documentation agent isn’t a separate demo, it’s powering the chat experience on this very site (look below at “Ask AI”). Every query you run against these docs uses Contextual AI’s Agent Composer capabilities to retrieve relevant information and generate accurate, grounded answers.

How It Works

  1. Your Documentation as Data: These docs are ingested into a Contextual AI datastore, chunked and indexed for retrieval. They’re continuously synced, so every time they are updated, the agent has access to the lastest info.
  2. Semantic Search: When you search, the agent retrieves the most relevant sections across all documentation pages.
  3. Grounded Generation: Answers are generated using retrieval-augmented generation (RAG), ensuring responses are grounded in the actual documentation—not hallucinated. You can check the citations for yourself.

Try It

Use the “Ask AI” bar below to ask a question about Contextual AI. You’ll see the documentation agent in action.

Ask a Question

Try searching the docs

Build Your Own

Want to create a documentation agent for your own content?