Skip to main content

Managing Chatbots

List, update, and organize your chatbots.

List Chatbots​

View all your created chatbots.

Via Web Interface​

  1. Navigate to Chatbots in the sidebar
  2. View all chatbots with:
    • Name and type
    • Model used
    • RAG status
    • Creation date
    • Last activity

Via API​

import requests

response = requests.get(
    "http://localhost/api/v1/chatbot/list",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

chatbots = response.json()["chatbots"]
for bot in chatbots:
    print(f"ID: {bot['id']}")
    print(f"Name: {bot['name']}")
    print(f"Model: {bot['model']}")
    print(f"RAG Enabled: {bot.get('use_rag', False)}")
    print(f"Created: {bot['created_at']}")
    print("-" * 40)

Get Chatbot Details​

Retrieve full configuration for a specific chatbot.

chatbot_id = "your-chatbot-id"

response = requests.get(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

chatbot = response.json()
print(f"Name: {chatbot['name']}")
print(f"System Prompt: {chatbot['system_prompt']}")
print(f"Temperature: {chatbot['temperature']}")
print(f"RAG Collection: {chatbot.get('rag_collection', 'N/A')}")

Update Chatbot​

Modify an existing chatbot's configuration.

Update Personality​

chatbot_id = "your-chatbot-id"

response = requests.put(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "system_prompt": "You are now a senior customer support agent. Provide detailed, technical solutions.",
        "temperature": 0.2,  # Lower temperature for more focused responses
        "fallback_responses": [
            "Let me research that for you.",
            "I'll connect you with our technical team."
        ]
    }
)

print(f"Updated chatbot: {response.json()['name']}")

Update RAG Settings​

response = requests.put(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "use_rag": True,
        "rag_collection": "updated_docs_collection",
        "rag_top_k": 7,  # Increase for more context
        "similarity_threshold": 0.6
    }
)

Change Model​

response = requests.put(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "gpt-4"  # Upgrade to more capable model
    }
)

Advanced Updates​

response = requests.put(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "max_tokens": 2000,
        "memory_length": 30
    }
)

Delete Chatbot​

Remove a chatbot you no longer need.

Via Web Interface​

  1. Navigate to Chatbots
  2. Find the chatbot to delete
  3. Click Delete
  4. Confirm deletion

Via API​

chatbot_id = "your-chatbot-id"

response = requests.delete(
    f"http://localhost/api/v1/chatbot/{chatbot_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

if response.status_code == 204:
    print("Chatbot deleted successfully")
else:
    print(f"Error: {response.json()}")

Clone Chatbot​

Create a new chatbot based on an existing one.

# Get original chatbot
original_id = "original-chatbot-id"

get_response = requests.get(
    f"http://localhost/api/v1/chatbot/{original_id}",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

original = get_response.json()

# Create new chatbot with same config
create_response = requests.post(
    "http://localhost/api/v1/chatbot/create",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "name": f"{original['name']} - Copy",
        "model": original["model"],
        "system_prompt": original["system_prompt"],
        "use_rag": original.get("use_rag", False),
        "rag_collection": original.get("rag_collection"),
        "temperature": original["temperature"]
    }
)

new_chatbot = create_response.json()
print(f"Created copy: {new_chatbot['id']}")

Batch Operations​

Manage multiple chatbots efficiently.

List and Filter​

response = requests.get(
    "http://localhost/api/v1/chatbot/list",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    params={
        "model": "gpt-3.5-turbo",  # Filter by model
        "use_rag": True  # Only chatbots with RAG enabled
    }
)

rag_chatbots = response.json()["chatbots"]
print(f"Found {len(rag_chatbots)} RAG-enabled chatbots")

Update Multiple​

import requests

chatbot_ids = ["chatbot-1", "chatbot-2", "chatbot-3"]

for chatbot_id in chatbot_ids:
    response = requests.put(
        f"http://localhost/api/v1/chatbot/{chatbot_id}",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        json={
            "temperature": 0.4,  # Update all to be more focused
            "memory_length": 15
        }
    )
    print(f"Updated {chatbot_id}: {response.status_code}")

Monitoring​

Track chatbot performance and usage.

Check Usage​

chatbot_id = "your-chatbot-id"

response = requests.get(
    f"http://localhost/api/v1/chatbot/{chatbot_id}/stats",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

stats = response.json()
print(f"Total Messages: {stats['total_messages']}")
print(f"Total Tokens: {stats['total_tokens']}")
print(f"RAG Queries: {stats['rag_queries']}")
print(f"Average Response Time: {stats['avg_response_time']}ms")

Check Performance​

response = requests.get(
    "http://localhost/api/v1/analytics/chatbot-performance",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    params={
        "chatbot_id": chatbot_id,
        "period": "7d"
    }
)

performance = response.json()
print(f"Success Rate: {performance['success_rate']:.2%}")
print(f"Average Response: {performance['avg_response_time']}ms")
print(f"RAG Hit Rate: {performance['rag_hit_rate']:.2%}")

Best Practices​

Organization​

  1. Use Descriptive Names - Make chatbots easy to identify
  2. Version Your Chatbots - Include version in names (e.g., "Support Bot v2")
  3. Document Changes - Keep track of configuration changes
  4. Test Changes - Test chatbots after major updates
  5. Monitor Performance - Check usage metrics regularly

Lifecycle Management​

StageAction
DevelopmentCreate and test extensively
StagingDeploy to test environment
ProductionMonitor and optimize
DeprecationMark as deprecated, plan migration

Security​

  1. Use Separate API Keys - Different keys for different chatbots
  2. Set Appropriate Permissions - Limit access per chatbot
  3. Regular Review - Review and clean up unused chatbots
  4. Audit Usage - Monitor which chatbots are used most

Troubleshooting​

Chatbot Not Responding​

Problem: Chatbot returns errors or no response

Check:

  • Chatbot ID is correct
  • Chatbot is enabled
  • API key has chatbot permissions
  • Model is available

RAG Not Working​

Problem: Chatbot doesn't use RAG documents

Check:

  • use_rag is true in configuration
  • RAG collection exists
  • Documents are uploaded to collection
  • Collection name matches configuration

Update Failing​

Problem: Update request returns error

Check:

  • You have permission to update the chatbot
  • JSON payload is valid
  • All required fields are included

Next Steps​