infoset-mcp
Provides MCP tools for accessing and managing Infoset CRM tickets, contacts, companies, email threads, SLA data, and ticket statistics.
List, search by subject, get, create, and update tickets with status, priority, owner, date, sort, pagination, stage, and pipeline options.
View ticket activity logs, add private or customer-visible HTML notes, and delete log entries.
Retrieve ticket stats, SLA breach data, email thread content, contacts, and companies.
List/search contacts by name, email, or phone.
Batch-fetch tickets, ticket logs, contacts, and companies in parallel, with deduplication for contacts and companies.
Supports API-key auth, rate-limit monitoring, and retry/backoff handling.
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., "@infoset-mcp show me open tickets from last week"
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.
infoset-mcp
Model Context Protocol server that gives AI assistants structured access to Infoset CRM tickets, contacts and companies.
Features
Wraps the Infoset CRM API into 18 MCP tools: tickets, contacts, companies, SLA, email threads and batch operations.
API key auth: stateless
X-API-Keyheader, no token refresh, no login flow.Rate limit aware: monitors
x-ratelimit-remaining, pauses 60s when remaining drops below 2.Retry with backoff: a 429 response triggers up to 3 retries, 60s apart.
Batch tools: fetch multiple tickets, logs, contacts or companies in one call, 5 requests in parallel, with deduplication for contacts and companies.
3 dependencies:
@modelcontextprotocol/sdk,axios,zod.
Related MCP server: GHL MCP Server
Installation
git clone https://github.com/sudohakan/infoset-mcp.git
cd infoset-mcp
npm install
cp .env.example .env # set INFOSET_API_KEYRegister it in your MCP client. The env-wrapper reads credentials from .env, so no secrets live in the client config:
{
"mcpServers": {
"infoset": {
"command": "/absolute/path/to/infoset-mcp/scripts/env-wrapper.sh",
"args": ["/absolute/path/to/infoset-mcp/src/mcp-server.mjs"],
"env": {}
}
}
}Configuration
Variable | Required | Default | Description |
| Yes | none | Infoset API key (sent as |
| No |
| API base URL |
| No | none | Your Infoset user ID, the default owner for |
Usage
Ticket operations
Tool | Description |
| List tickets with status, owner, date, sort filters and pagination |
| Get full ticket detail by ID |
| Search tickets by keyword with status and priority filters |
| Create a ticket with subject, contact, priority, content |
| Update status, priority, owners, subject, stage, pipeline |
| Get activity logs for a ticket |
| Add a note to the timeline, private by default, HTML allowed |
| Delete a log entry, such as a note added by mistake |
| Ticket count breakdown by status |
Contacts, companies, email and SLA
Tool | Description |
| Get contact details by ID |
| List or search contacts by name, email, phone |
| Get company details by ID |
| Get email thread content by email ID |
| Get SLA breach data for a ticket |
Batch operations
Tool | Description |
| Get multiple tickets in parallel (concurrency 5) |
| Get logs for multiple tickets in parallel |
| Get multiple contacts with auto-deduplication |
| Get multiple companies with auto-deduplication |
Reference codes
1 | 2 | 3 | 4 | |
Status | Open | Pending | Resolved | Closed |
Priority | Low | Normal | High | Urgent |
Development
npm test # test suite
npm run test:coverage # with coverage report
npm start # start the server standaloneSee CONTRIBUTING.md for code style and pull request guidelines, and SECURITY.md for credential handling and vulnerability reporting.
Project structure
src/mcp-server.mjs single-file server: tools, API client, rate limit monitor, retry handler
scripts/env-wrapper.sh loads .env, then starts the server
tests/ Jest suite (mcp-server.test.mjs)MCP client (Claude, Cursor) <--stdio--> infoset-mcp (Node.js) <--HTTPS--> Infoset APITransport is stdio (spawned by the MCP client, JSON over stdin/stdout, logs to stderr). Every request carries the X-API-Key header.
License
MIT, see LICENSE.
Available Tools
18 toolsinfoset_add_ticket_noteA
Add a note to a ticket timeline (POST /v1/tickets/logs). Private by default (agents only); set isPrivate=false to make it visible to the customer. Notes accept HTML. Returns the created log, whose id can be passed to infoset_delete_ticket_log.
Inputs: notes (required), ticketId (required), isPrivate.
infoset_batch_get_companiesB
Get multiple Infoset companies in a single call (auto-deduplicates)
Inputs: companyIds (required).
infoset_batch_get_contactsB
Get multiple Infoset contacts in a single call (auto-deduplicates)
Inputs: contactIds (required).
infoset_batch_get_ticket_logsB
Get activity logs for multiple Infoset tickets in a single call
Inputs: ticketIds (required), itemsPerPage.
infoset_batch_get_ticketsB
Get multiple Infoset tickets in a single call (parallel fetch)
Inputs: ticketIds (required).
infoset_create_ticketB
Create a new Infoset ticket
Inputs: subject (required), contactId (required), status, content, +1 more.
infoset_delete_ticket_logA
Delete a ticket log entry, such as a note added by mistake (DELETE /v1/tickets/logs/{logId}). Irreversible.
Inputs: logId (required).
infoset_get_companyB
Get company information by company ID
Inputs: companyId (required).
infoset_get_contactA
Get contact information by contact ID
Inputs: contactId (required).
infoset_get_emailB
Get email thread content by email ID
Inputs: emailId (required).
infoset_get_sla_breachesA
Get SLA breach data for an Infoset ticket
Inputs: ticketId (required).
infoset_get_ticketA
Get single Infoset ticket detail by ID
Inputs: ticketId (required).
infoset_get_ticket_logsC
Get activity logs for an Infoset ticket
Inputs: ticketId (required), sortDir, itemsPerPage.
infoset_get_ticket_statsC
Get ticket statistics dashboard (counts by status)
Inputs: toDate, fromDate, ownerIds.
infoset_list_contactsB
List or search Infoset contacts
Inputs: page, query, itemsPerPage.
infoset_list_ticketsC
List Infoset tickets with optional filters and pagination
Inputs: page, status, sortCol, sortDir, +3 more.
infoset_search_ticketsA
Search Infoset tickets by subject text (contains, case-insensitive) and optional filters. The API has no full-text search: Search/q/keyword are ignored and return every ticket. Company lookup is not exposed as a tool: over REST use companies?name= then tickets?companyId=.
Inputs: query (required), page, status, priority, +1 more.
infoset_update_ticketC
Update an existing Infoset ticket
Inputs: ticketId (required), status, stageId, subject, +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 18 tools
Most tools target a distinct resource+action, and the single/batch/log variants are clearly separated by name. The main fuzzy boundary is infoset_list_tickets vs infoset_search_tickets, which the description itself admits is only a subject-contains filter with no real full-text search. Minor asymmetry: notes are created with infoset_add_ticket_note but removed with infoset_delete_ticket_log, which an agent must infer.
Every tool follows the infoset_<verb>_<noun> pattern, including the uniform batch_get_* group, list_*, get_*, create_*, update_*, add_*, and delete_*. No camelCase/snake_case mixing or stray vague verbs; the convention is predictable enough to guess tool names.
18 tools across tickets, logs, contacts, companies, email, stats, and SLA is reasonable for a helpdesk API, but the four batch_get_* tools largely duplicate existing single-get capabilities and push the surface toward the heavy side. Each tool still has a defensible purpose, so this is slightly over-scoped rather than bloated.
Ticket lifecycle is well covered (create, update, get, list, search, notes, logs, stats), but the contact and company side is read-only — no create/update for contacts or companies, and no company list/lookup tool at all (the description explicitly says company lookup is not exposed). Emails are only reachable by ID via infoset_get_email with no list/discovery path, and there is no ticket deletion, leaving several dead ends.
Maintenance
Related MCP Connectors
MCP server enabling AI agents to manage Bitrix24 features via standardized protocol
MCP server unifying ERPs, CRMs, APIs and knowledge base for Claude, ChatGPT and Gemini.
API-first CRM for LLMs - contacts, companies, deals and activities over a native MCP server.
MCP server for LeadDelta — manage LinkedIn connections and CRM data via AI assistants.
Related MCP Servers
- AlicenseAqualityBmaintenanceMCP server for Creatio CRM that enables AI assistants to read, create, update, delete records, execute business processes, and manage settings via OData and REST APIs.2456 npm6MIT
- AlicenseBqualityDmaintenanceA comprehensive MCP server that connects AI assistants to GoHighLevel CRM, enabling management of contacts, conversations, calendars, pipelines, payments, and more through 60+ tools.6437 npm1MIT
- AlicenseNot gradedqualityDmaintenanceMCP server for amoCRM integration, enabling AI assistants to manage leads, contacts, companies, tasks, and more via natural language.4MIT
- AlicenseNot gradedqualityBmaintenanceMCP server for integrating AI assistants with Bitrix24, enabling retrieval and management of CRM entities, tasks, scrum, and knowledge bases through natural language.1MIT