Customer Support Chatbot using RAG

Overview

This tutorial demonstrates how to:

Set up a web crawler using Julep’s Spider integration
Process and store crawled content in a document store
Implement RAG for enhanced AI responses
Create an intelligent agent that can answer questions about crawled content

Task Structure

Let’s break down the task into its core components:

1. Input Schema

First, we define what inputs our task expects:

Copy
input_schema:
    type: object
    properties:
        url:
            type: string
        pages_limit:
            type: integer

This schema specifies that our task expects:

A URL string (e.g., “https://julep.ai/”)
Number of pages to crawl (e.g., 5)

2. Tools Configuration

Next, we define the external tools our task will use:

Copy
- name: spider_crawler
  type: integration
  integration:
    provider: spider
    method: crawl
    setup:
      spider_api_key: YOUR_SPIDER_API_KEY

- name: create_agent_doc
  type: system
  system:
    resource: agent
    subresource: doc
    operation: create

We’re using two tools:

The spider_crawler integration for web crawling
The create_agent_doc system tool for storing processed content

3. Main Workflow Steps

Crawl Website

Copy
- tool: spider_crawler
  arguments:
    url: $ _['url']
    params:
      request: "smart_mode"
      limit: $ _['pages_limit']
      return_format: "markdown"
      proxy_enabled: "True"
      filter_output_images: "True"
      filter_output_svg: "True"
      readability: "True"
      sitemap: "True"
      chunking_alg:
        type: "bysentence"
        value: "15"

This step:

Takes the input URL and crawls the website
Processes content into readable markdown format
Chunks content into manageable segments
Filters out unnecessary elements like images and SVGs

Process and Index Content

Copy
- evaluate:
    documents: $ _['content']

- over: $ "[(steps[0].input.content, chunk) for chunk in _['documents']]"
  parallelism: 3
  map:
    prompt:
    - role: user
      content: >-
        $ f'''
        <document>
        {_[0]}
        </document>

        <chunk>
        {_[1]}
        </chunk>

        Please give a short succinct context to situate this chunk within the overall document.'''

This step:

Processes each content chunk in parallel
Generates contextual metadata for improved retrieval
Prepares content for storage

Store Documents

Copy
- over: $ _['final_chunks']
  parallelism: 3
  map:
    tool: create_agent_doc
    arguments:
      agent_id: $ agent.id
      data:
        metadata:
          source: spider_crawler
        title: Website Document
        content: $ _

This step:

Stores processed content in the document store
Adds metadata for source tracking
Creates searchable documents for RAG

YAML
Copy
# yaml-language-server: $schema=https://raw.githubusercontent.com/julep-ai/julep/refs/heads/dev/schemas/create_task_request.json
name: Julep Customer Support RAG Task
description: A Julep agent that can crawl a website and store the content in the document store.

########################################################
################### INPUT SCHEMA #######################
########################################################

input_schema:
  type: object
  properties:
    url:
      type: string
    pages_limit:
      type: integer
      
########################################################
################### TOOLS ##############################
########################################################

tools:
- name: spider_crawler
  type: integration
  integration:
    provider: spider
    method: crawl
    setup:
      spider_api_key: "YOUR_SPIDER_API_KEY"

- name : create_agent_doc
  description: Create an agent doc
  type: system
  system:
    resource: agent
    subresource: doc
    operation: create

########################################################
################### INDEX PAGE SUBWORKFLOW ##############
########################################################

index_page:

# Step #0 - Evaluate the content
- evaluate:
    documents: $ _['content']

# Step #1 - Process each content chunk in parallel
- over: $ "[(steps[0].input.content, chunk) for chunk in _['documents']]"
  parallelism: 3
  map:
    prompt: 
    - role: user
      content: >-
        $ f'''
        <document>
        {_[0]}
        </document>

        Here is the chunk we want to situate within the whole document
        <chunk>
        {_[1]}
        </chunk>

        Please give a short succinct context to situate this chunk within the overall document for the purposes of improving search retrieval of the chunk. 
        Answer only with the succinct context and nothing else.'''
    
    unwrap: true
    settings:
      max_tokens: 16000

# Step #2 - Create a new document and add it to the agent docs store
- evaluate:
    final_chunks: |
      $ [
        NEWLINE.join([chunk, succint]) for chunk, succint in zip(steps[1].input.documents, _)
      ]

# Step #3 - Create a new document and add it to the agent docs store
- over: $ _['final_chunks']
  parallelism: 3
  map:
    tool: create_agent_doc
    arguments:
      agent_id: "00000000-0000-0000-0000-000000000000" # <--- This is the agent id of the agent you want to add the document to
      data:
        metadata:
          source: "spider_crawler"

        title: "Website Document"
        content: $ _

########################################################
################### MAIN WORKFLOW ######################
########################################################

main:

# Step #0 - Define a tool call step that calls the spider_crawler tool with the url input
- tool: spider_crawler
  arguments:
    url: $ _['url'] # You can also use 'steps[0].input['url']'
    params:
      request: "smart_mode"
      limit: $ _['pages_limit'] # <--- This is the number of pages to crawl (taken from the input of the task)
      return_format: "markdown"
      proxy_enabled: "True"
      filter_output_images: "True" # <--- This is to execlude images from the output
      filter_output_svg: "True" # <--- This is to execlude svg from the output
      # filter_output_main_only: "True"
      readability: "True" # <--- This is to make the output more readable
      sitemap: "True" # <--- This is to crawl the sitemap
      chunking_alg: # <--- Using spider's bysentence algorithm to chunk the output
        type: "bysentence"
        value: "15" # <--- This is the number of sentences per chunk

# Step #1 - Evaluate step to document chunks
- foreach:
    in: $ _['result']
    do:
      workflow: index_page
      arguments:
        content: $ _.content

Usage

Start by creating an execution for the task. This execution will make the agent crawl the website and store the content in the document store.

Copy
from julep import Client
import time
import yaml

# Initialize the client
client = Client(api_key=JULEP_API_KEY)

# Create the agent
agent = client.agents.create(
  name="Julep Crawler Agent",
  description="A Julep agent that can crawl a website and store the content in the document store.",
)

# Load the task definition
with open('crawling_task.yaml', 'r') as file:
  task_definition = yaml.safe_load(file)

# Create the task
task = client.tasks.create(
  agent_id=agent.id,
  **task_definition
)

# Create the execution
execution = client.executions.create(
  task_id=task.id,
  input={"url": "https://julep.ai/", "pages_limit": 5}
)
# Wait for the execution to complete
while (result := client.executions.get(execution.id)).status not in ['succeeded', 'failed']:
    print(result.status)
    time.sleep(1)

if result.status == "succeeded":
    print(result.output)
else:
    print(f"Error: {result.error}")

Next, create a session for the agent. This session will be used to chat with the agent.

Copy
session = client.sessions.create(
  agent_id=AGENT_ID
)

Finally, chat with the agent.

Copy
response = client.sessions.chat(
  session_id=session.id,
  messages=[
    {
      "role": "user",
      "content": "What is Julep?"
    }
  ]
)

print(response)

Example Output

This is an example output when the agent is asked “What is Julep?”

Next Steps

Try this task yourself, check out the full example, see the Customer Support Chatbot using RAG cookbook.

Get Started

Julep CLI

Core Concepts

Tutorials

Integrations

Advanced Topics

Customer Support Chatbot using RAG

Overview

Task Structure

1. Input Schema

2. Tools Configuration

3. Main Workflow Steps

Usage

Example Output

Next Steps

Get Started

Julep CLI

Core Concepts

Tutorials

Integrations

Advanced Topics

​Overview

​Task Structure

​1. Input Schema

​2. Tools Configuration

​3. Main Workflow Steps

​Usage

​Example Output

​Next Steps

Overview

Task Structure

1. Input Schema

2. Tools Configuration

3. Main Workflow Steps

Usage

Example Output

Next Steps