Customer Support

By Connic

AI ticket triage with sentiment analysis and RAG-powered response drafting. Classifies by priority, blocks spam, and stores new solutions.

supportbeginnerWebSocket HTTP Webhook Email

View on GitHub Quickstart

connic init my-project --templates=customer-support

Overview

Classifies incoming tickets by priority, category, and customer sentiment, then drafts empathetic responses using semantic search over your knowledge base. Middleware blocks spam senders with StopProcessing before the LLM runs, saving tokens. New solutions discovered during response drafting are stored back into the knowledge base for future tickets. Tools use context injection for run traceability.

Use cases

Real-time chat support

Connect via WebSocket for streaming responses in a live chat widget with sentiment-aware tone matching.

Ticketing system integration

Process tickets from Zendesk or Intercom via webhook, auto-classify and suggest first responses.

Knowledge base building

Automatically grow your support knowledge base as agents discover and store new solutions.

Architecture

WebSocket / Webhook

support-pipeline

support-triager

support-responder

Knowledge Base

Scaffolded project structure

Running connic init my-project --templates=customer-support creates this file tree.

customer-support/
  agents/
    support-triager.yaml
    support-responder.yaml
    support-pipeline.yaml
  tools/
    support_tools.py
  middleware/
    support-triager.py
  schemas/
    ticket-classification.json
  requirements.txt
  README.md

Get started

Install the template, create a Connic project, and deploy. Choose Git (automatic on push) or CLI (works with any provider).

Prerequisites

Python 3.10+
A Connic account (create a project first)
API key for your LLM provider (e.g. Gemini, OpenAI) to add in project variables

Create project

Install and scaffold

Install the SDK and create a project from this template.

terminal

pip install connic-composer-sdk
connic init my-project --templates=customer-support

Then cd my-project

Deploy

Pick your deployment method. Git auto-deploys on push; CLI works with GitLab, Bitbucket, or no Git.

Git integration

In Connic: Project Settings → Git & Environments, connect your GitHub repo
On the same page, map your branch (e.g. main) to the Production environment
Push your scaffolded project to that repo

terminal

git add .
git commit -m "Add Customer Support template"
git push origin main

CLI deploy

In Connic: Project Settings → CLI, create an API key and copy project ID
Run connic login in your project folder
connic dev to try with hot-reload, connic test to run suites, or connic deploy for production

terminal

connic login
connic dev     # Ephemeral dev env with hot-reload
connic test    # Run declarative test suites
connic deploy  # Deploy to production

Connect and configure

Add a WebSocket connector for real-time chat with streaming responses. Optionally add an HTTP Webhook (outbound) connector to forward escalations to Slack or PagerDuty. Add your LLM provider API key in Project Settings → LLM Provider.

Full connector setup guide Quickstart guide Connector docs

Template source

Browse the full template, contribute improvements, or fork for your own use.

connic-org/connic-awesome-agents/tree/main/customer-support

Information

Publisher: By Connic
Category: Agent Templates
Difficulty: beginner
Works with: WebSocket, HTTP Webhook, Email
Source: GitHub

More agent templates

Browse the marketplace

Telegram Personal Assistant

Multimodal Telegram assistant that processes text, images, voice, video, and documents. Searches the web, saves notes and knowledge, and schedules follow-up reminders with persistent sessions.

S3 Document Pipeline

Auto-process documents uploaded to S3 with text extraction, classification, and routing. Handles PDFs, images, and text files with retries.

SQS Order Processor

Validate and fulfill orders from SQS queues with fraud detection, inventory checks, and concurrency keys. Results publish to an outbound queue.

Email Helpdesk

End-to-end email support with IMAP polling, auto-reply filtering, and intent classification. Drafts contextual replies from your knowledge base.

Invoice Processor

Extract structured data from invoices with AI, validate totals via a deterministic tool agent, and output auditable JSON with retry support.

Kafka Fraud Detector

Score transactions for fraud in real-time on Kafka streams. Uses velocity analysis, geo-anomaly detection, and pattern matching per customer.