Free RAG Chatbot
Ask this RAG chatbot anything about Fetch Hive. It searches a knowledge base built from our docs, answers from what it finds and shows the exact passages and pages behind every answer. When the docs donโt cover something, it says so.
Chat with the Fetch Hive docs
A Fetch Hive agent with one tool, Vector Search, and one knowledge base, our own docs. Ask it anything about Fetch Hive. Every answer shows the passages it came from and the pages they sit on.
An example chat from a real run. Ask your own question below to start a live one.
How many tasks does a knowledge base search cost, and can Vector Search run on a schedule every morning?
A Search Knowledge Base call costs 1 fixed platform task per request. The docs I found do not say that Vector Search can be scheduled to run every morning, so I can't confirm that from the docs.
Passages the answer came from
โSearch Knowledge Base | 1 | Fixed platform task credit cost.โ
โ- Build support or ops agents that search a curated Knowledge Base. - Retrieve source material before the agent writes a response.โ
Source pages
Try a question
Want this chatbot on your own docs?
This is a Fetch Hive agent, not a black box. Clone it, attach a knowledge base built from your own docs, files or site, and put it behind your help centre, your app or Slack.
Prefer chat? Run it from Claude or ChatGPT, free as a guest for 72 hours.
An Answer You Can Check, Not Just Read
Most AI chatbots sound sure whether theyโre right or not. A RAG chatbot is different: it retrieves passages from a knowledge base first, then writes the answer from them. This one shows you those passages, so you can see where every answer came from.
A chatbot without retrieval
0
sources shown with the answer
Answers from whatever the model remembers
No way to check where a claim came from
A confident answer even when itโs wrong
What this RAG chatbot shows you
Up to 4
quoted passages behind each answer
Answers written only from the knowledge base
Every passage quoted word for word, with its page
What the docs donโt cover is said, not guessed
The knowledge base here is the Fetch Hive docs. Clone the agent to put the same chatbot on your own content.
What Each Answer Includes
Everything comes from passages the knowledge base search returned for that question. A quote that isnโt in those results is dropped before you see it.
Output 1
A Short Answer
Two to six plain sentences, or a short list of steps, written only from the passages the search found.
Returns
Answer ยท follow-up questions
Output 2
The Passages It Used
Up to four quotes, copied word for word from the knowledge base and checked against the search results.
Returns
Quote ยท checked against the search
Output 3
The Source Pages
Each passage links to the docs page and section it sits on, so you can open it and read the rest.
Returns
Page title ยท section ยท link
Output 4
What the Docs Donโt Cover
When the knowledge base only answers part of the question, the chatbot answers that part and tells you what is missing.
Returns
Not covered note ยท run receipt
How This RAG Chatbot Works
The chat above is one Fetch Hive agent with one tool, Vector Search, and one knowledge base. For every question it searches the knowledge base, reads the closest passages and writes the answer from them. Then each quote is checked against what the search returned. Every run is logged.
See what every step does and what it costs
Every run of this workflow in Fetch Hive logs all of this, so you know what happened and what it cost.
- Every prompt
- Every model call
- Every tool call
- Every workflow step
- Every input and output
- Every cost
Clone it, then swap the knowledge base
Attach a knowledge base built from your own docs, help centre, files or site. Change the tone, the model or the rules. Itโs your copy, so nothing is locked.
Keep the knowledge base current
Add or replace sources as your content changes, so the chatbot answers from the latest version and not last yearโs.
Call it from your own app
Every agent can be called through the Fetch Hive API, so your site, app or Slack bot can send a question and get the same JSON back.
What This RAG Chatbot Can and Cannot Do
It answers from the Fetch Hive docs only. Ask it about anything else and it will tell you the docs donโt cover it.
A chat holds up to six questions. After that, start a new chat.
It uses the earlier questions in the chat for context, so follow-ups like โand what does that cost?โ work.
It only says what the docs say. If a page is out of date, the answer will be too.
It can still misread a passage. Thatโs why the quotes and pages sit next to every answer.
How to Use the RAG Chatbot
Ask a question the way youโd ask a colleague, then follow up. Hereโs what people build once they clone it.
Chat with your docs
Point the agent at your documentation and let your team ask it questions instead of searching.
Automate it in Fetch Hive
Attach your docs as a knowledge base and share the agent.
Answer customers on your site
Put an AI chatbot with a knowledge base behind your help widget, so answers come from your help centre and nothing else.
Automate it in Fetch Hive
Call the agent from your site through the API.
Give your team an internal assistant
Load policies, runbooks and onboarding notes, and let people ask in Slack instead of pinging a colleague.
Automate it in Fetch Hive
Connect the agent to Slack and answer in the thread.
Find the gaps in your content
Every โnot coveredโ answer is a question your docs donโt answer yet. Collect them and you have a writing list.
Automate it in Fetch Hive
Log unanswered questions to a sheet for your docs team.
Put This RAG Chatbot on Your Own Knowledge Base
Clone the agent behind this chat, attach a knowledge base built from your content and connect it to where questions arrive. Youโll see every search, every answer and what each one costs.
Docs chatbot
On each question
New question
A question arrives from your site, app or Slack
Search
Find the closest passages in your knowledge base
Answer
Reply with the answer, quotes and source pages
Run the RAG chatbot right inside Claude
Connect Fetch Hive once, then ask for this report in the chat you already have open. Free for 72 hours as a guest, no account needed.
Which AI assistant do you use?
Open Connectors
In Claude (web or Desktop), open Settings, then Connectors, and click Add custom connector.
Paste the connector URL
Name it Fetch Hive and paste this URL.
https://connect.fetchhive.com/mcpConnect as a guest
Click Connect. When Fetch Hive opens in your browser, choose Continue as guest (or sign in), then click Allow.
Then ask Claude
Search my docs knowledge base and answer this with quoted sources: How do I rotate an API key?
Create a free account any time to keep what your assistant built. Stuck? Read the setup guide.
Frequently asked questions
What a RAG chatbot is, what this free one shows and where its limits are.
A RAG chatbot answers questions by first retrieving passages from a knowledge base, then writing its answer from those passages. RAG stands for retrieval-augmented generation. Because the answer is tied to real text, it can show where it came from, which is what this one does.
Still have questions?
More free tools
Explore More Free Tools
Want answers from other sources too? Draft support replies from a help centre, pull structured data from any page and research a company in one run.
Customer Service Response Generator
Paste your help centre URL and a customer question, and get a draft reply with a source on every claim.
Built withTry for freeFree ยท No signupAI Web Scraper
Paste a URL and get the page back as a clean table, JSON and markdown.
Built withTry for freeFree ยท No signupCompany Research Tool
Enter a companyโs website and get a short account brief with a source on every claim.
Built withTry for freeFree ยท No signupAI PDF Data Extractor
Choose a PDF or paste the text and get the fields and line items back as a table, CSV and JSON.
Built with
Try for freeFree ยท No signup





