NextCRM MCP Server

Community

Open-source CRM MCP server with 127 tools across 15 modules for AI agent access.

How to install ↓ Compare GitHub source repository ↗

Category: Other #205 of 230
Stars: ★ 718 Very popular
Transport: Streamable HTTP · SSE
Runtime: Hosted (no local runtime)
Credentials: API key / credential required
License: MIT
Last commit: today
Tools: 80

35FMRS · D

The NextCRM MCP server offers comprehensive CRM data access for AI agents, with rich feature coverage and support for multiple transports. It includes detailed configuration instructions and is open-source, though it remains self-hosted and not officially endorsed. Overall, it is a valuable tool for teams using NextCRM who want to leverage AI automation.

Strongest · Security and permissions 8/20 Weakest · Documentation 6/20

Reliability

8/20

Security and permissions

8/20

Maintenance

7/20

Documentation

6/20

Setup experience

6/20

Why each score

Reliability 8/20: Static review only; runtime behavior not verified. The source code mentions 127 tools across 15 modules, but no executable tests or CI configuration proving this were found. The MCP server is implemented via mcp-handler and requires credentials to verify. Therefore, reliability is capped at 12.
Security and permissions 8/20: Tokens are stored as SHA-256 hashes and API keys encrypted with AES-256-GCM, which is good practice. However, MCP tools include delete operations and sensitive actions like sending emails, but no explicit confirmation mechanism is evident. Although external connections are present by default, no malicious code is observed. Score 8 because credential handling and encryption are solid, but explicit confirmation for destructive operations is missing.
Maintenance 7/20: The repository is active with 673 stars, but stars do not add points. MIT license is clear, but no security response channel or explicit versioning is provided. Commits and issue responses are not evidenced in the source. Development is active, but governance and release management have gaps.
Documentation 6/20: The README is comprehensive, covering installation and Docker setup. However, MCP tool parameters, limitations, and troubleshooting are not well documented. There is no dedicated MCP documentation, only a mention of a SKILL.md. Setup instructions assume prior knowledge of external services. Documentation is usable but missing key details.
Setup experience 6/20: Setup steps are clear with Docker Compose and manual instructions. However, MCP server setup requires generating tokens and configuring the client, and dependencies need additional configuration. No CI or tests are present to verify the setup works. Score is limited because of lack of automated verification.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkConnects to a database

Best for

Developers and teams using NextCRM who want AI agents to access their CRM data directly
Users who want to combine their CRM operations with AI tools like Claude Code or Cursor
Organizations seeking to automate CRM data enrichment and campaign management

Not for

Users not using NextCRM as their underlying CRM system
Teams without the technical capacity to deploy and maintain a NextCRM instance
Users requiring a fully production-grade MCP server backed by a dedicated vendor

Required permissions

The MCP server uses Bearer token authentication; you need to generate a token of the form nxtc__...
Tokens are generated from the profile page and stored SHA-256 hashed; the raw value is shown only once
AI agents can perform read and write operations, including creating, updating, and deleting CRM records

Risks and side effects

API keys (e.g., OpenAI, Anthropic) could be misused if not managed carefully; a 3-tier priority system with encrypted storage is recommended
Granting AI agents write access via the MCP server could lead to unintended data modifications; use with caution
Bearer tokens, if leaked, pose a security risk; store them securely.
Dependency on third-party services (E2B, Resend, Inngest) that may become unavailable or incur costs

Setup

Before you start

Runtime：Hosted (no local runtime)

Deploy the NextCRM instance first (see README for installation or Docker Compose). After logging in, generate a Bearer token (format: nxtc_...) from your profile page's developer tab. Then add the JSON configuration to your MCP client (see install_config). Ensure to replace the URL with your actual NextCRM domain.

claude_desktop_config.json

{
  "mcpServers": {
    "nextcrm": {
      "type": "http",
      "url": "https://your-nextcrm.com/api/mcp/mcp",
      "headers": {
        "Authorization": "Bearer nxtc__your_token_here"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "nextcrm": {
      "type": "http",
      "url": "https://your-nextcrm.com/api/mcp/mcp",
      "headers": {
        "Authorization": "Bearer nxtc__your_token_here"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add --transport http nextcrm https://your-nextcrm.com/api/mcp/mcp --header 'Authorization: Bearer nxtc__your_token_here'

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After connecting, the client's tool list should show the nextcrm server's tools (e.g. list_accounts, list_contacts). Ask "List my CRM accounts" — returned account data confirms the endpoint and Bearer token are configured correctly.

Troubleshooting

Ensure the NextCRM app is correctly deployed and running.
Check the MCP client configuration for a correct URL (/api/mcp/mcp or /api/mcp/sse).
Verify the Bearer token is valid and not expired.
Ensure environment variables (e.g., OPENAI_API_KEY) are set, as some features rely on AI integrations.
Check NextCRM logs and monitoring (e.g., Inngest) for background job errors

Things to try

Once connected, you can ask your AI assistant things like:

List all accounts in my CRM.
Search contacts with name 'Novak' and show their details.
Create a new opportunity worth 50,000 linked to the existing account.
Show me the stats for last month's email campaign.

Tools 80

list_accounts read-only

List account records

get_account read-only

Get a single account

search_accounts read-only

Search account records

create_account writes

Create an account

update_account writes

Update an account

delete_account destructive

Delete an account

list_contacts read-only

List contact records

get_contact read-only

Get a single contact

Show 72 more tools

search_contacts read-only

Search contact records

create_contact writes

Create a contact

update_contact writes

Update a contact

delete_contact destructive

Delete a contact

list_leads read-only

List lead records

get_lead read-only

Get a single lead

search_leads read-only

Search lead records

create_lead writes

Create a lead

update_lead writes

Update a lead

delete_lead destructive

Delete a lead

list_opportunities read-only

List opportunity records

get_opportunity read-only

Get a single opportunity

search_opportunities read-only

Search opportunity records

create_opportunity writes

Create an opportunity

update_opportunity writes

Update an opportunity

delete_opportunity destructive

Delete an opportunity

list_targets read-only

List target records

get_target read-only

Get a single target

search_targets read-only

Search target records

create_target writes

Create a target

update_target writes

Update a target

delete_target destructive

Delete a target

list_products read-only

List products

get_product read-only

Get a single product

create_product writes

Create a product

update_product writes

Update a product

delete_product destructive

Delete a product

list_contracts read-only

List contracts

get_contract read-only

Get a single contract

create_contract writes

Create a contract

update_contract writes

Update a contract

delete_contract destructive

Delete a contract

list_activities read-only

List activities

get_activity read-only

Get a single activity

create_activity writes

Create an activity

update_activity writes

Update an activity

delete_activity destructive

Delete an activity

list_documents read-only

List documents

get_document read-only

Get document details

create_document writes

Create a document

upload_document writes

Upload a document

download_document read-only

Download a document

link_document writes

Link a document to an entity

unlink_document writes

Unlink a document from an entity

delete_document destructive

Delete a document

list_target_lists read-only

List target lists

get_target_list read-only

Get target list details

create_target_list writes

Create a target list

update_target_list writes

Update a target list

delete_target_list destructive

Delete a target list

add_members writes

Add members to a target list

remove_members writes

Remove members from a target list

enrich_contact writes

Enrich contact data using AI

enrich_target writes

Enrich target data using AI

bulk_enrich_contacts writes

Bulk enrich contacts

bulk_enrich_targets writes

Bulk enrich targets

list_email_accounts read-only

List email accounts

campaign_crud writes

Perform CRUD operations on campaigns

send_campaign destructive

Send a campaign

pause_campaign writes

Pause a campaign

resume_campaign writes

Resume a campaign

campaign_templates read-only

Manage campaign templates

campaign_steps read-only

Manage campaign steps

campaign_stats read-only

Get campaign statistics

project_boards read-only

Manage project boards

project_sections writes

Manage project sections

project_tasks writes

Manage project tasks

project_comments writes

Manage project comments

project_documents read-only

Manage project documents

project_watch writes

Watch or unwatch a project

list_reports read-only

List available reports

run_report read-only

Run a report

Use cases

Enable AI agents to manage accounts, contacts, and opportunities automatically

Query and update CRM data through natural language

Automate sales target enrichment and campaign management

Integrate with custom AI workflows for reporting and project collaboration

Supported clients

Claude Desktop

Cursor

Listed from the project's documentation, not tested by this site.

Overview

NextCRM MCP server is a built-in Model Context Protocol (MCP) server of the NextCRM open-source CRM, enabling AI agents (Claude, Cursor, custom agents) to read and write CRM data directly. It provides 127 tools across 15 modules including Accounts, Contacts, Leads, Opportunities, Targets, Products, Contracts, Activities, Documents, Target Lists, Enrichment, Email Accounts, Campaigns, Projects, and Reports. The server supports both streamable HTTP (recommended) and legacy SSE transports, with Bearer token authentication. Tokens are generated from the profile page and stored SHA-256 hashed.

Similar servers

Hevy MCP 77 · B

Manage your Hevy workout data from AI assistants via MCP.

★ 509 · Tools 26 Compare with this →

Firecrawl MCP Server 75 · B

Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents

★ 7.6k · Tools 11 Compare with this →

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 654 · Tools 11 Compare with this →

Compartment 75 · B

Encrypted, fully offline long-term memory for AI agents

★ 579 · Tools 14 Compare with this →

Compare with other MCP servers

Source revision 98c4d17ffd9b Data synced 2026-10-12 Read the FMRS scoring method