Cinode

by Unknown

Not rated
Website

About

Give agents a view of people, projects, skills and assignments

Details

Author
Unknown
Categories
Other, Project Management

The Cinode MCP (Model Context Protocol) server securely connects AI clients and MCP-compatible agents to your Cinode environment so they can access your organisation’s live data. That gives agents the same contextual view of people, projects, skills and assignments that your team has, instantly, so they can assist, analyse, update and automate with confidence.

Think of it as giving your AI assistant a working knowledge of your Cinode setup, without exposing data you don’t want shared.

- Smarter AI support– Agents act with real context from your Cinode data
- Faster workflows– Reduce time spent on manual updates and lookups
- Always in sync– Agents work with live, up-to-date information
- Secure by design– Fine‑grained access control over what agents can see and change

You can access our MCP server with Streamable HTTP athttps://mcp.cinode.com/_mcp

You need to be on a paid plan using Cinode to have access to our API and our MCP Server.

Our MCP server supportstwo authentication approaches:
- OAuth (recommended): Browser-based sign-in using Dynamic Client Registration (DCR). No manual token handling.
- Bearer Token: Static API token authentication
- Basic Credentials: Client ID and Client Secret authentication

Bearer tokens and Basic credentials are sensitive information.

Never share or store them insecurely (for example: in screenshots, public/shared documents, ticket attachments, chat tools like Teams/Slack, emails, sticky notes, or websites). Always use secure storage and only transmit credentials over encrypted connections (HTTPS/TLS).

With OAuth you don't create or paste any token. Supported MCP clients connect to the server URL and handle authentication for you:

Basic authentication uses your Client ID and Client Secret:
- Join them with a colon:CLIENT_ID:CLIENT_SECRET
- Base64-encode the full string.

When you generate a Personal API Account in Cinode we'll show you an encoded and ready-to-use string which you immediately can use within theAuthorizationheader. Read morehere.

The examples below are generic templates.Always refer to your provider's official documentationfor the most up-to-date configuration instructions, as setup details may vary between versions and providers.

{ "mcpServers": { "cinode": { "command": "npx", "args": [ "mcp-remote", "https://mcp.cinode.com/_mcp", "--header", "Authorization: ${AUTH_HEADER}" ], "env": { "AUTH_HEADER": "Bearer YOUR_CINODE_API_TOKEN" } } } }

Note:Replace"npx"above with"C:\\Progra~1\\nodejs\\npx.cmd"if you're running Windows.

- Claude Desktop:MCP setup documentationor follow steps below:

- Settings > Connectors > Add custom connector, pointing athttps://mcp.cinode.com/_mcp
- You only need to fill inNameandServer URL.

Using MCP together with AI makes it possible to performlarge-scale and automated changesto your Cinode data. While powerful, this also increases the risk of unintended or irreversible outcomes if prompts are incorrect or poorly tested.

When write access is enabled, AI-driven actions can modify or delete large amounts of data in a short time.

You are responsible for how MCP and AI tools are used in your organisation.
Changes made through MCP may bepermanentand cannot always be undone.
Use MCP write capabilitiesat your own risk.

To reduce risk, Cinode strongly recommends that you:

- Only enable write tools when necessaryand when you are confident in both the technology and your prompt capabilities.
- Validate prompts using read-only accessbefore allowing create, update, or delete operations.
- Limit write accessto experienced users who understand Cinode’s data model and business impact.
- Review and monitor automated workflows regularly.
- Make sure your provider's privacy settings are set according to internal policy.

The Cinode MCP server does not send your data anywhere on its own. It only responds to the specific requests made by the AI client you connect - and only within the permissions of the account that authenticated.

- When you connect an AI client (for exampleClaude/Anthropic, ChatGPT/OpenAI, or Microsoft Copilot), that client - not Cinode - is the party that processes your prompts and the data it retrieves.
- Only the data an agent actually queries is returned to it and passed to your chosen AI provider. There is no bulk export or background syncing of your Cinode database.
- Every request is scoped to the connected user's existing Cinode roles and permissions. An agent can never see or change more than that user already can.

- How the retrieved data is stored, retained, or used for model training is governed byyour chosen AI provider's terms and privacy settings, not by Cinode. Review and configure them according to your internal policy.
- Any data processing agreement (DPA) for AI processing is between your organisation and the AI provider - not with Cinode.
- Only enable write tools once you are confident in your prompts and permissions (seeSecurity & Governanceabove).

By default, employees can createpersonal API accountsand use those credentials (for example, Basic tokens) to connect MCP clients and AI tools to Cinode. This means that MCP access does not have to be centrally provisioned by an administrator.

- A personal API accountcannot be used to do morethan the user can already do in the Cinode application.
- All actions performed through MCP are limited by the user’s existing roles and permissions.
- MCP access follows the same permission boundaries as normal interactive use of Cinode.

For organisations that want tighter control, administrators can:

- Disable personal API account creation, preventing users from connecting MCP clients on their own.
-
Revoke or rotate existing API tokensat any time.
-
Adjust user roles and permissions, which immediately affects what MCP-connected tools can access or change.

MCP server for Tripletex — Norwegian ERP and accounting platform. Invoices, customers, employees, projects, and ledger entries.

AI-native trade business management for Australian tradies. 11 MCP tools for leads, quotes, jobs, invoicing, financials, and client CRM.

Australian trade business OS with 67 MCP tools across 23 domains — autonomous business formation, lead management, quoting, job scheduling, invoicing, crypto payments, and growth analytics for tradies.

Interact with and manage your Bitrix24 CRM instance through a powerful set of tools.

The Bitrix24 MCP Server is designed to connect external systems to Bitrix24. It provides AI agents with standardized access to Bitrix24 features and data via the Model Context Protocol (MCP). The MCP server enables external AI systems to interact with Bitrix24 modules through a single standardized interface. You can connect the Bitrix24 MCP Server to the AI model you already use and manage Bitrix24 directly from it. The MCP server allows actions to be performed and data to be retrieved strictly within the access rights configured in your Bitrix24: the AI agent receives only the information and capabilities that are explicitly requested and authorized. Interaction with the Tasks module is supported (the list of supported modules and available actions is gradually expanding).

Integrate Odoo with Large Language Models (LLMs) for enhanced business process automation.

Interact with Odoo ERP systems, allowing AI assistants to access and manage business data like contacts, sales, and projects.

Integrate with Odoo ERP systems to manage business data and operations. Requires external configuration for connection parameters.

A hosted or self-hosted cloud version of Odoo MCP

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.