Allows managing Facebook Messenger contacts, conversations, and messages via the DMTicket omnichannel platform.
Allows managing Telegram contacts, conversations, and messages via the DMTicket omnichannel platform.
Allows managing Zalo contacts, conversations, and messages via the DMTicket omnichannel platform.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@DMTicket MCP Server show me all my VIP contacts"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Model Context Protocol server for DMTicket — the omnichannel customer-messaging & CRM platform (Zalo, Facebook, Telegram, groups & communities).
It gives AI agents (Claude, Cursor, ChatGPT, custom agents…) live access to a DMTicket account: contacts, conversations, messages, labels, custom attributes, inboxes, teams, agents, canned responses, automation and reports — so an agent can read, segment, message and automate on your behalf.
Most helpdesk integrations stop at "list my conversations". DMTicket exposes the pieces an agent needs to stand up a working business process on its own:
Data sources — register a connection to a Lark sheet, PostgreSQL, or MySQL. The agent references it by name; the credentials stay server-side and are never handed to the model. Reads only — SQL that is not a single SELECT is rejected.
Scheduled flows — automation that fires on a cron schedule rather than waiting for a conversation to exist. Read rows, filter them, extract fields, skip what was already handled, then draft or send.
Dry run — POST /scheduled_flows/{id}/run defaults to a dry run: the pipeline executes for real, but actions are reported instead of performed. An agent can show you exactly what a flow will send before it sends anything.
Durable dedupe — automation_states claims a key atomically, so a restart, a redeploy, or an overlapping run never processes the same row twice.
Hand an agent your token and this server, describe the process in plain language, and it can build and test the automation end to end.
Related MCP server: DLF Agency MCP Server
On startup the server fetches the DMTicket OpenAPI spec from https://dmticket.com/api/openapi.json and auto-generates one MCP tool per API operation — 129 tools at the time of writing. When DMTicket adds a new endpoint, it becomes an MCP tool on the next restart, with no code changes here.
Point it somewhere else with DMTICKET_OPENAPI_URL if you self-host.
Sign in to your DMTicket account at https://app.dmticket.com
Click your avatar (bottom-left) → Profile settings
Scroll to the bottom and copy the Access Token
The token carries your own permissions, so an agent using it can do exactly what you can — no more. Keep it secret; regenerate from the same screen to revoke.
API base URL:
https://app.dmticket.com· Auth header:api-access-token: <token>Quick test:curl https://app.dmticket.com/api/v1/profile -H "api-access-token: <token>"Spell the header with dashes. Rails maps
api-access-tokenandapi_access_tokento the same value, but proxies strip headers containing underscores, so the underscored spelling arrives empty and every call 401s.
Most endpoints are scoped to one account, so you also need your numeric account id — it is the number in the app URL, e.g. app.dmticket.com/app/accounts/**3**/dashboard.
git clone https://github.com/doanhnd9989/dmticket-mcp.git
cd dmticket-mcp
npm install
npm run build # outputs dist/Set two environment variables:
Variable | Required | Default | Description |
| ✅ | — | Your workspace token from step 1 |
| — |
| Your DMTicket origin |
| — |
|
|
claude mcp add dmticket \
-e DMTICKET_API_KEY=<your-workspace-token> \
-e DMTICKET_API_URL=https://dmticket.com \
-- node /absolute/path/to/dmticket-mcp/dist/index.mjsOr in claude_desktop_config.json / .mcp.json:
{
"mcpServers": {
"dmticket": {
"command": "node",
"args": ["/absolute/path/to/dmticket-mcp/dist/index.mjs"],
"env": {
"DMTICKET_API_KEY": "<your-workspace-token>",
"DMTICKET_API_URL": "https://dmticket.com"
}
}
}
}.cursor/mcp.json){
"mcpServers": {
"dmticket": {
"command": "node",
"args": ["/absolute/path/to/dmticket-mcp/dist/index.mjs"],
"env": { "DMTICKET_API_KEY": "<your-workspace-token>" }
}
}
}DMTICKET_API_KEY=<token> DMTICKET_MCP_TRANSPORT=sse DMTICKET_MCP_PORT=3333 node dist/index.mjs
# then point your agent at http://<host>:3333/sseTool names are the OpenAPI operationId in snake_case — e.g. list_contacts, create_contact, list_conversations, create_message, create_data_source, query_data_source, create_scheduled_flow, run_scheduled_flow, list_scheduled_flow_runs, claim_automation_state, create_webhook, … The full set is generated live from the spec, so it always matches your DMTicket version.
The token is a per-workspace credential — treat it like a password. Do not commit it.
The server only talks to your configured DMTICKET_API_URL.
Revoke by regenerating the workspace token in DMTicket settings.
ISC. This is the MCP integration for DMTicket (https://dmticket.com).
Add a New Agent
Inputs: no named parameters.
Add a New Agent
Inputs: inbox_id (required), user_ids (required).
Add a New Agent
Inputs: user_ids (required).
Add a new article
Inputs: no named parameters.
Add a new automation rule
Inputs: no named parameters.
Add a New Canned Response
Inputs: no named parameters.
Add a new category
Inputs: no named parameters.
Add a new custom attribute
Inputs: no named parameters.
Add a new portal
Inputs: no named parameters.
Assign Conversation
Inputs: team_id, assignee_id.
Claim a key before doing irreversible work
Inputs: no named parameters.
Get Contactable Inboxes
Inputs: id (required).
Add Labels
Inputs: labels (required).
Contact Conversations
Inputs: id (required).
Create Contact
Inputs: no named parameters.
Delete Contact
Inputs: no named parameters.
Show Contact
Inputs: no named parameters.
Contact Filter
Inputs: page, payload.
Create contact inbox
Inputs: id (required), inbox_id (required), source_id.
List Contacts
Inputs: no named parameters.
Merge Contacts
Inputs: base_contact_id (required), mergee_contact_id (required).
Search Contacts
Inputs: q.
Update Contact
Inputs: no named parameters.
Add Labels
Inputs: labels (required).
Conversations Filter
Inputs: page, payload.
Conversations List
Inputs: q, page, labels, status, +3 more.
Get Conversation Counts
Inputs: q, labels, status, team_id, +1 more.
Create a contact
Inputs: no named parameters.
Create a conversation
Inputs: no named parameters.
Create a custom filter
Inputs: no named parameters.
Create a label
Inputs: no named parameters.
Create a message
Inputs: no named parameters.
Create an Agent Bot
Inputs: no named parameters.
Create New Message
Inputs: no named parameters.
Create an integration hook
Inputs: no named parameters.
Create a team
Inputs: no named parameters.
Add a webhook
Inputs: no named parameters.
Create a data source
Inputs: no named parameters.
Create a scheduled flow
Inputs: no named parameters.
Tạo chính sách SLA
Inputs: sla_policy.
Delete a custom filter
Inputs: no named parameters.
Remove an Agent from Account
Inputs: id (required).
Remove an Agent from Inbox
Inputs: inbox_id (required), user_ids (required).
Remove an Agent from Team
Inputs: user_ids (required).
Delete a label
Inputs: no named parameters.
Delete a message
Inputs: no named parameters.
Delete an AgentBot
Inputs: no named parameters.
Delete an Integration Hook
Inputs: no named parameters.
Delete a team
Inputs: no named parameters.
Remove a automation rule from account
Inputs: id (required).
Delete an automation state entry
Inputs: namespace (required).
Delete a webhook
Inputs: no named parameters.
Remove a Canned Response from Account
Inputs: id (required).
Remove a custom attribute from account
Inputs: id (required).
Delete a data source
Inputs: no named parameters.
Delete a scheduled flow
Inputs: no named parameters.
Xoá chính sách SLA
Inputs: no named parameters.
Fetch user profile
Inputs: no named parameters.
List Agents in Account
Inputs: no named parameters.
List Audit Logs in Account
Inputs: page.
List all automation rules in an account
Inputs: no named parameters.
Get account branded email layout
Inputs: no named parameters.
List all Canned Responses in an Account
Inputs: no named parameters.
Account Conversation Metrics
Inputs: no named parameters.
List all custom attributes in an account
Inputs: attribute_model (required).
Get account details
Inputs: no named parameters.
Account Reporting Events
Inputs: name, since, until, user_id, +1 more.
Agent Conversation Metrics
Inputs: no named parameters.
Get conversation statistics grouped by agent
Inputs: no named parameters.
Get one automation state entry
Inputs: namespace (required).
Get conversation statistics grouped by channel type
Inputs: no named parameters.
Get messages from a conversation
Inputs: no named parameters.
Conversation Reporting Events
Inputs: no named parameters.
Get CSAT survey page
Inputs: no named parameters.
Get a data source
Inputs: no named parameters.
Get a contact
Inputs: no named parameters.
Conversation Details
Inputs: no named parameters.
Inbox details
Inputs: no named parameters.
List all the Integrations
Inputs: no named parameters.
Get an agent bot details
Inputs: no named parameters.
Get a automation rule details
Inputs: id (required).
Get a custom attribute details
Inputs: id (required).
Get a custom filter details
Inputs: no named parameters.
Get a label
Inputs: no named parameters.
Get a team details
Inputs: no named parameters.
Get first response time distribution by channel
Inputs: no named parameters.
Get an inbox
Inputs: id (required).
Show Inbox Agent Bot
Inputs: id (required).
Get inbox-label matrix report
Inputs: no named parameters.
List Agents in Inbox
Inputs: no named parameters.
Get conversation statistics grouped by inbox
Inputs: no named parameters.
Báo cáo hội thoại theo nhãn (Type)
Inputs: no named parameters.
Get outgoing messages count grouped by entity
Inputs: group_by (required).
List all portals in an account
Inputs: no named parameters.
Get a scheduled flow
Inputs: no named parameters.
Get a single conversation
Inputs: no named parameters.
Xem một chính sách SLA
Inputs: no named parameters.
List Agents in Team
Inputs: no named parameters.
Get conversation statistics grouped by team
Inputs: no named parameters.
Create an inbox
Inputs: no named parameters.
List all AgentBots
Inputs: no named parameters.
List all conversations
Inputs: no named parameters.
List all messages
Inputs: no named parameters.
Get Account reports
Inputs: no named parameters.
Get Account reports summary
Inputs: no named parameters.
List all custom filters
Inputs: no named parameters.
List all inboxes
Inputs: no named parameters.
List all labels
Inputs: no named parameters.
List Labels
Inputs: no named parameters.
List Labels
Inputs: no named parameters.
Get messages
Inputs: after, before.
List all teams
Inputs: no named parameters.
List all webhooks
Inputs: no named parameters.
List automation state entries
Inputs: namespace (required), key, limit.
Danh sách đánh giá mức độ hài lòng (CSAT)
Inputs: no named parameters.
List all data sources
Inputs: no named parameters.
List past runs of a flow
Inputs: limit.
List scheduled flows
Inputs: no named parameters.
Danh sách chính sách SLA
Inputs: no named parameters.
Create New Conversation
Inputs: no named parameters.
Write an automation state entry
Inputs: no named parameters.
Read from the data source
Inputs: no named parameters.
Resolve a conversation
Inputs: no named parameters.
Run a flow now (dry run by default)
Inputs: no named parameters.
Test the connection
Inputs: no named parameters.
Toggle Priority
Inputs: priority (required).
Toggle Status
Inputs: status (required), snoozed_until.
Toggle typing status
Inputs: typing_status (required).
Toggle Typing Status
Inputs: typing_status (required), is_private.
Update account
Inputs: no named parameters.
Update account branded email layout
Inputs: no named parameters.
Update a contact
Inputs: no named parameters.
Update a custom filter
Inputs: no named parameters.
Add or remove agent bot
Inputs: id (required), agent_bot (required).
Update Agent in Account
Inputs: id (required).
Update Agents in Inbox
Inputs: inbox_id (required), user_ids (required).
Update Agents in Team
Inputs: user_ids (required).
Update a label
Inputs: no named parameters.
Update a message
Inputs: no named parameters.
Update an agent bot
Inputs: no named parameters.
Update an Integration Hook
Inputs: no named parameters.
Update a team
Inputs: no named parameters.
Update automation rule in Account
Inputs: id (required).
Update a webhook object
Inputs: no named parameters.
Update Canned Response in Account
Inputs: id (required).
Update Conversation
Inputs: priority, sla_policy_id.
Update custom attribute in Account
Inputs: id (required).
Update Custom Attributes
Inputs: custom_attributes (required).
Update a data source
Inputs: no named parameters.
Update Inbox
Inputs: id (required).
Update last seen
Inputs: no named parameters.
Update a portal
Inputs: no named parameters.
Update user profile
Inputs: profile (required).
Update a scheduled flow
Inputs: no named parameters.
Sửa chính sách SLA
Inputs: sla_policy.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 155 tools
Many tools have overlapping purposes, e.g., 'create_a_contact' and 'contact_create' both create contacts. The naming is inconsistent, making it hard to distinguish between similar operations.
Tool names follow no consistent pattern, mixing 'contact_list', 'create_a_contact', 'get_details_of_a_contact', and 'contact_create' without rhyme or reason.
With 155 tools, the server is far too large for a typical MCP tool set. This excessive granularity undermines usability and suggests many tools could be merged or parameterized.
While the server covers many entities (contacts, conversations, agents, etc.), the massive redundancy and poor organization indicate significant gaps in coherent workflow coverage.
Email, WhatsApp and Telegram for AI agents: send, campaigns, automations, contacts, agent inboxes.
Work your HelpRabbit help desk: tickets, replies, customers, the AI agent and its knowledge.
Read and answer support conversations, and run outbound email and chat sequences, from your AI.
Send email, manage domains, templates, webhooks and broadcasts from your AI agent.
