Blocks
Customer Service Agent
Specialized customer support agent
Create a specialized customer service agent that can handle support tickets, analyze sentiment, categorize issues, and provide appropriate responses with escalation recommendations.
Block Preview
Customer Service Agent
Specialized customer support agent
Usage
- Add the block to your workflow and connect it to the upstream step.
- Configure any required credentials or tokens in the inputs.
- Fill in required inputs and optional parameters for the run.
- Run a test execution, inspect outputs, and iterate before deploying.
- Deploy the customer_service_agent block with monitoring enabled in production.
Inputs (UI)
System Prompt
long-inputPlaceholder: You are a professional customer service agent. Always be helpful, empathetic, and solution-focused...
Layout: full
Customer Message
long-inputPlaceholder: Enter the customer message or support ticket...
Layout: full
Customer Context
long-inputPlaceholder: Enter customer history, account info, previous interactions...
Layout: full
Model
dropdownLayout: half
Options: Dynamic options
Temperature
sliderLayout: half
API Key
short-inputPlaceholder: Enter your API key
Layout: full
Condition: model = ["qwen2.5:0.5b","qwen2.5:1.5b","qwen2.5:3b","llama3.2:1b","llama3.2:3b","phi3.5:3.8b"]
Knowledge Base
long-inputPlaceholder: Enter company policies, FAQ, product information...
Layout: full
Knowledge Sources
knowledge-source-inputLayout: full
Max Search Results
sliderLayout: half
Condition: knowledgeSources = ""
Similarity Threshold
sliderLayout: half
Condition: knowledgeSources = ""
Escalation Rules
codePlaceholder: { "keywords": ["refund", "cancel", "angry", "lawsuit", "complaint"], "sentimentThreshold": "negative", "priorityThreshold": "high", "autoEscalate": true }
Layout: full
Agent Type
dropdownLayout: half
Options: None (Default), Chatbot, Customer Assistant, HR Agent
Response Style
dropdownLayout: half
Options: Professional, Friendly, Empathetic, Concise
Include Analysis
switchLayout: half
Tools
tool-inputLayout: full
Response Format
codePlaceholder: { "type": "object", "properties": { "content": {"type": "string"}, "sentiment": {"type": "string", "enum": ["positive", "neutral", "negative"]}, "priority": {"type": "string", "enum": ["low", "medium", "high", "urgent"]}, "category": {"type": "string"}, "suggestedActions": {"type": "array", "items": {"type": "string"}}, "escalationRequired": {"type": "boolean"} } }
Layout: full
Enable Memory
switchLayout: half
Memory Limit
sliderLayout: half
Condition: memoryEnabled = true
Importance Threshold
sliderLayout: half
Condition: memoryEnabled = true
Memory Tags
short-inputPlaceholder: customer-support, billing, technical
Layout: half
Condition: memoryEnabled = true
Inputs (API)
systemPrompt
stringOptional
customerMessage
stringRequired
customerContext
stringOptional
model
stringRequired
apiKey
stringRequired
temperature
numberOptional
agentType
stringOptional
knowledgeBase
stringOptional
knowledgeSources
stringOptional
searchMaxResults
numberOptional
similarityThreshold
numberOptional
escalationRules
jsonOptional
responseStyle
stringOptional
includeAnalysis
booleanOptional
tools
jsonOptional
responseFormat
jsonOptional
memoryEnabled
booleanOptional
memoryLimit
numberOptional
memoryImportance
numberOptional
memoryTags
stringOptional
Outputs
Primary response type:
{
"content": "string",
"model": "string",
"sentiment": "string",
"priority": "string",
"category": "string",
"suggestedActions": "json",
"escalationRequired": "boolean",
"tokens": "json"
}Conditional output based on: responseFormat
When empty: {"content":"string","model":"string","sentiment":"string","priority":"string","category":"string","suggestedActions":"json","escalationRequired":"boolean","tokens":"json"} | When filled: "json"