MCP Server¶

The Daktela MCP Server is a read-only Model Context Protocol (MCP) server for the Daktela Contact Centre REST API v6. It lets AI assistants such as Claude, ChatGPT, and other LLM-based clients access and analyse your contact centre data directly.

The MCP server is hosted directly by your Daktela instance — no installation, container, or self-hosting is required. You connect to it over HTTP and authenticate with a Daktela API Access Token.

What is MCP?¶

The Model Context Protocol is an open standard that lets AI applications connect to external data sources and tools. By connecting your AI assistant to the Daktela MCP server, you give it structured, read-only access to your Daktela instance — enabling tasks like email quality audits, sales pipeline reviews, and call analysis without writing any code.

Info

The MCP server provides read-only access to your Daktela data. It cannot create, modify, or delete any records in your instance.

Requirements¶

A Daktela instance with the MCP server enabled
An MCP-compatible client (e.g. Claude Code, Claude Desktop)
A Daktela API Access Token

Tip

You can generate an API Access Token in your Daktela instance under Manage > Users > Edit user > Access Tokens.

Setup¶

There are two equivalent ways to connect your client to the Daktela MCP server. Pick whichever fits your workflow — a per-project configuration or a global configuration that applies to all your projects.

In both cases, replace YOUR_DOMAIN with your Daktela hostname (e.g. your-instance.daktela.com) and YOUR_ACCESS_TOKEN with your API Access Token.

Project configuration (`.mcp.json`)¶

Place the following in your project's .mcp.json file:

{
  "mcpServers": {
    "daktela": {
      "type": "http",
      "url": "https://YOUR_DOMAIN/mcp-server/mcp",
      "headers": {
        "X-Daktela-Access-Token": "YOUR_ACCESS_TOKEN"
      }
    }
  }
}

Claude Code global settings (`~/.claude/settings.json`)¶

Alternatively, add the same block to your user's global ~/.claude/settings.json so the connection is available across all your projects:

{
  "mcpServers": {
    "daktela": {
      "type": "http",
      "url": "https://YOUR_DOMAIN/mcp-server/mcp",
      "headers": {
        "X-Daktela-Access-Token": "YOUR_ACCESS_TOKEN"
      }
    }
  }
}

Available Tools¶

The MCP server exposes 43 read-only tools organised into the following categories. All list tools support pagination, sorting, and contextual filtering.

Tickets¶

Tool	Description
`count_tickets`	Count tickets matching filter criteria
`get_ticket`	Get a specific ticket by ID
`get_ticket_detail`	Get detailed ticket information
`list_account_tickets`	List tickets for a specific account
`list_ticket_categories`	List available ticket categories
`list_tickets`	List tickets with filtering and pagination

Activities¶

Tool	Description
`count_activities`	Count activities matching filter criteria
`get_activity`	Get a specific activity by ID
`list_activities`	List activities with filtering and pagination

Calls¶

Tool	Description
`count_calls`	Count calls matching filter criteria
`get_call`	Get a specific call by ID
`get_call_transcript`	Get the transcript of a specific call
`list_call_transcripts`	List available call transcripts
`list_calls`	List calls with filtering and pagination

Emails¶

Tool	Description
`count_emails`	Count emails matching filter criteria
`get_email`	Get a specific email by ID
`list_emails`	List emails with filtering and pagination

Messaging¶

Covers webchat, SMS, Facebook Messenger, Instagram, WhatsApp, and Viber.

Tool	Description
`count_chats`	Count chat messages matching filter criteria
`get_chat`	Get a specific chat message by ID
`list_chats`	List chat messages with filtering and pagination

Contacts and CRM¶

Tool	Description
`count_accounts`	Count accounts matching filter criteria
`count_contacts`	Count contacts matching filter criteria
`count_crm_records`	Count CRM records matching filter criteria
`get_account`	Get a specific account by ID
`get_contact`	Get a specific contact by ID
`get_crm_record`	Get a specific CRM record by ID
`list_accounts`	List accounts with filtering and pagination
`list_contacts`	List contacts with filtering and pagination
`list_crm_records`	List CRM records with filtering and pagination

Campaigns¶

Tool	Description
`count_campaign_records`	Count campaign records matching filter criteria
`get_campaign_record`	Get a specific campaign record by ID
`list_campaign_records`	List campaign records with filtering and pagination
`list_campaign_types`	List available campaign types

Reference Data¶

Tool	Description
`list_groups`	List agent groups
`list_pauses`	List available pause types
`list_queues`	List queues
`list_statuses`	List ticket/activity statuses
`list_templates`	List message templates
`list_users`	List users

Real-time¶

Tool	Description
`list_realtime_sessions`	List current real-time agent sessions

Knowledge Base¶

Tool	Description
`list_article_folders`	List knowledge base folders
`list_articles`	List knowledge base articles
`get_article`	Get a specific knowledge base article

Built-in Prompts¶

The server includes four pre-built prompt templates for common analysis workflows. These prompts are available directly in your AI client when the MCP server is connected.

Prompt	Description
`email_quality_audit`	Audit recent emails for negative sentiment, unprofessional tone, and lost deals
`sales_pipeline_review`	Review deal health and identify at-risk opportunities
`call_quality_review`	Analyse calls for escalations, knowledge gaps, and quality issues
`daily_call_analysis`	Daily call review for churn risk and recurring issues

Usage Examples¶

Once the MCP server is connected to your AI client, you can ask questions like:

"Show me all open tickets from the last 7 days and summarise common issues."
"Audit the last 50 emails for negative sentiment and unprofessional tone."
"List all missed calls from today and identify patterns."
"Review the sales pipeline and flag deals that have been stalled for more than 2 weeks."
"Analyse call transcripts from this week and highlight any escalations."

The AI assistant will use the appropriate MCP tools to fetch data from your Daktela instance and provide structured analysis.

Best Practices¶

Token security: Never commit .mcp.json or settings.json containing real API Access Tokens to public repositories. Treat the token like a password.
Dedicated API user: Create a dedicated API user with minimal read-only permissions for MCP access, rather than reusing an administrator's token. This limits the data exposed to the AI client and makes it easier to revoke access if needed.

Support¶

If you need help connecting to the MCP server or have questions about its behaviour, feel free to contact our support.