CAS Genesis World MCP
About
An MCP server for the CAS genesisWorld REST Webservice v7.0.
Details
- Author
- gnidreve
- Categories
- Database
Jump to
Setup
Install CAS Genesis World MCP in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/gnidreve/Genesis-World-MCP
Follow the installation instructions in the repository README, then restart your MCP client.
Connect Claude, Cursor, or anyMCP-compatible AI agent to yourCAS genesisWorldCRM. Search contacts, manage tasks and appointments, and read or write records — through natural language, without writing a single API call.
- 69 tools, including7 native flowsthat bundle multi-step CRM operations (like checking for scheduling conflicts before booking a meeting, or checking for duplicates before creating a contact) into a single call.
- Full read/write accessto tasks, contacts, appointments, documents, distribution lists, and more — plus generic access to any custom object type your installation defines.
- Read-only modeavailable for safe, exploratory use.
- Ships as a ready-madeDocker imageornpm package.
Add this to your MCP client's config (e.g. Claude Desktop'sclaude_desktop_config.json):
{ "mcpServers": { "cas-genesisworld": { "command": "npx", "args": ["-y", "cas-genesis-world-mcp"], "env": { "GENESISWORLD_BASE_URL": "http://your-genesisworld-server/genesisrest.svc", "GENESISWORLD_PRODUCT_KEY": "your-product-key", "GENESISWORLD_USERNAME": "your-username", "GENESISWORLD_PASSWORD": "your-password" } } } }
Restart your client — the tools show up automatically. Ask your agent to find a contact, list your open tasks, or book a meeting, and it takes it from there.
Prefer a persistent, self-hosted server instead of a per-client process? SeeSelf-hosting with Dockerbelow.
Once connected, your agent can handle requests like:
- "Find the contact info for Jane Doe and show me her open tasks."
- "Create a follow-up task for this lead and link it to their contact record."
- "Book a meeting with the sales team next Tuesday at 10am — check everyone's calendar for conflicts first."
- "Check for possible duplicates before creating a new contact for Acme Corp."
- "Generate a report for this opportunity."
Requests like these are answered byflows— single tool calls that bundle the multi-step sequence of API requests a human would otherwise have to script by hand.
Write (23 — hidden in read-only mode, along with the write flows above)
The full upstream API this is built on is committed asswagger.json.
Beyond tools, the server exposes MCP resources for data that rarely changes, so your agent doesn't burn tool calls rediscovering it every session:
- genesisworld://readme— static orientation document for the agent itself (domain model, navigation patterns, efficiency rules).
- genesisworld://types— data-object types accessible to the user, with permissions. Cached 15 min.
- genesisworld://metadata/{objectType}— field/relationship schema of one type, e.g.genesisworld://metadata/ADDRESS. Cached 15 min.
- genesisworld://views/{objectType}— saved views of one type, e.g.genesisworld://views/TASK. Cached 15 min.
For a persistent server multiple clients can point at (instead of onenpxprocess per client):
docker run -d --name cas-genesisworld-mcp -p 8084:3000 \ -e GENESISWORLD_BASE_URL="http://your-genesisworld-server/genesisrest.svc" \ -e GENESISWORLD_PRODUCT_KEY="your-product-key" \ -e GENESISWORLD_USERNAME="your-username" \ -e GENESISWORLD_PASSWORD="your-password" \ vaatu/cas-genesis-world-mcp # Read-only mode: append --read-only after the image name docker run -d --name cas-genesisworld-mcp -p 8084:3000 \ -e GENESISWORLD_BASE_URL="http://your-genesisworld-server/genesisrest.svc" \ -e GENESISWORLD_PRODUCT_KEY="your-product-key" \ -e GENESISWORLD_USERNAME="your-username" \ -e GENESISWORLD_PASSWORD="your-password" \ vaatu/cas-genesis-world-mcp --read-only
Multi-tenant: one server, many genesisWorld identities
By default the container has one fixed genesisWorld identity for every client that connects. With--client-credentials, it has none — each client authenticates itself, so one server can safely serve several people/teams with different genesisWorld logins:
docker run -d --name cas-genesisworld-mcp -p 8084:3000 \ -e GENESISWORLD_BASE_URL="http://your-genesisworld-server/genesisrest.svc" \ vaatu/cas-genesis-world-mcp --client-credentials
Each client then sends its own credentials as headers when connecting:
{ "mcpServers": { "cas-genesisworld": { "url": "http://localhost:8084/mcp", "headers": { "X-GenesisWorld-Username": "your-username", "X-GenesisWorld-Password": "your-password" } } } }
X-GenesisWorld-Product-Keyis optional here — if you keepGENESISWORLD_PRODUCT_KEYset on the server, clients that omit their own product key fall back to it. HTTP transport only (there's no per-request header channel on stdio); a request missing both username and password headers is rejected outright (HTTP 401), it never falls back to a shared identity.
The MCP endpoint is now athttp://localhost:8084/mcp. Point your client at it:
{ "mcpServers": { "cas-genesisworld": { "url": "http://localhost:8084/mcp" } } }
Two kinds of settings, kept strictly separate — no setting is both:Environmentsconfigure the deployment (where the API lives, who connects);launch optionstoggle behavior at startup.
- Required in practice for any real request to succeed — except in
--client-credentialsmode (see below), where the server has no identity of its own and none of these three are required.
docker run(as shown above). None of them have an environment-variable equivalent, by design.
Want to add a tool or understand how this is built? SeeAGENTS.mdfor architecture and contributor docs, andROADMAP.mdfor the project plan.
Official Airtable MCP server and skills for working with bases, records, workflows, and business operations from AI agents.
MCP Server For Apache Doris, an MPP-based real-time data warehouse.
Official MCP Server from Atlan which enables you to bring the power of metadata to your AI tools
Query Onchain data, like ERC20 tokens, transaction history, smart contract state.
Read and write access to your Baserow tables.
Introspect and query your apps deployed to Convex.
Interact with the data stored in Couchbase clusters using natural language.
Maritime intelligence for tracking vessels, analysing ports, and exploring ship data.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.



