genesys-mcp
MCP server for Genesys Cloud that enables AI assistants to build and manage queues, skills, users, wrap-up codes, and Architect flows, including a flow builder that renders diagrams and publishes via Genesys' validation pipeline.
README
genesys-mcp
Your Genesys Cloud org, in your AI's hands. An open-source MCP server for Genesys Cloud on Cloudflare Workers. Zero dependencies, no terminal required, and its whole purpose is to build: queues, skills, users, wrap-up codes, and Architect flows.
It builds, not just reads.
Prompt Claude (or any MCP client):
- "create a queue called Weekend Support"
- "build a call flow: greet callers, press 1 for Weekend Support, press 9 to end the call. show me the diagram first"
- "draw my Main IVR as a diagram"
- "onboard prep: create a spanish skill and assign it to Jess at proficiency 4"
The flow builder composes real Archy YAML, shows you the flow as a Mermaid diagram in chat, then publishes through Genesys' own flow-jobs pipeline (the same one CX as Code uses), so Genesys validates and publishes server-side.
What it deliberately does NOT do
- No analytics or KPI tools. That lane is already covered; for conversation analytics over MCP, check out MakingChatbots' genesys-cloud-mcp-server.
- No contact-list management.
- No deletes. There are no delete tools, and the raw API tool refuses
DELETE. Create-first by design.
Deploy your own in 3 steps
- Deploy: click the button (free Cloudflare account), or
git clone+npx wrangler deploy. The CONFIG KV namespace is auto-provisioned. - Create a Genesys OAuth client: Admin → Integrations → OAuth → Add Client → grant type Client Credentials → assign a role (see Scoping the role) → save.
- Configure: open
/setupon your new Worker and paste the Client ID, Secret, and your region. The wizard validates them live against Genesys before saving, then hands you your access key (shown once).
Prefer terminal-managed config? Set Wrangler secrets instead; they override the wizard: GENESYS_CLIENT_ID, GENESYS_CLIENT_SECRET, GENESYS_REGION (e.g. usw2.pure.cloud), MCP_AUTH_TOKEN.
Connect your AI
The MCP endpoint is https://<your-worker>/mcp.
- Claude (web/desktop): Settings → Connectors → Add custom connector → paste the URL. When the authorization screen appears, paste your access key.
- Claude Code:
claude mcp add --transport http genesys https://<your-worker>/mcpand authenticate when prompted. - ChatGPT: Settings → Connectors → Advanced → Developer mode → add the MCP server URL.
- Anything else: standard streamable HTTP MCP with OAuth 2.1 (or send the access key as a Bearer token).
Then try: "check the connection and list my queues."
The toolbox (25 tools)
| Group | Tools |
|---|---|
| 🔌 Org & Connection | about, check_connection, list_divisions, list_did_pools |
| 📞 Queues & Routing | list_queues, get_queue, create_queue ✏️, list_wrapup_codes, create_wrapup_code ✏️ |
| 👥 Users & Skills | list_users, get_user, list_skills, create_skill ✏️, assign_user_skill ✏️ |
| 🌳 Flows (Architect) | list_flows, get_flow, get_flow_configuration, list_prompts, render_flow, export_flow |
| 🏗️ Flow Builder | build_flow, publish_flow ✏️, get_flow_job, unlock_flow ✏️ |
| ⚡ Power | genesys_api_call ✏️ (any Platform API endpoint; GET/POST/PUT/PATCH only) |
✏️ = write tool. Everything resolves names to GUIDs for you, so "the Weekend Support queue" just works.
How the flow builder works
build_flowturns a spec (greeting + DTMF menu + queue transfers) into Archy YAML and a Mermaid diagram. Your AI shows you the diagram first.publish_flowregisters an Architect flow job, uploads the YAML, and polls. Genesys validates and publishes server-side; validation errors come back verbatim.render_flowalso diagrams flows that already exist in your org, andexport_flowround-trips any flow back to YAML.
TTS is inline in the flow (your org's TTS engine speaks it), so there are no audio files to record or upload.
Security model
- Your Genesys credentials live in your Cloudflare account (Wrangler secrets or the auto-provisioned KV), and nowhere else.
- The AI can only do what the OAuth client's role allows. You scope it; Genesys enforces it.
- The MCP endpoint requires the access key (raw Bearer or the built-in OAuth 2.1 flow for connectors).
- No delete tools exist, and
genesys_api_callrefusesDELETEoutright.
Scoping the role
For everything here to work, the OAuth client's role needs: Routing (queues, skills, wrap-up codes) view + add, Directory user view + edit (for skill assignment), Architect flow view + add + edit + publish, and Telephony DID view. Master Admin works for a sandbox; scope down for production. The server can only ever be as powerful as the role you assign.
Local dev
git clone https://github.com/ryanshatz/genesys-mcp
cd genesys-mcp
cp .dev.vars.example .dev.vars # or create it: GENESYS_CLIENT_ID / GENESYS_CLIENT_SECRET / GENESYS_REGION / MCP_AUTH_TOKEN
npm test # zero-dep unit tests (node --test)
npm run smoke # live read-tools smoke against your org
npm run smoke -- --writes # also exercises create tools (MCP_Test_* artifacts)
npm run dev # wrangler dev
About
Built in the open by Ryan Shatzkamer (Director, Technical Services at outboundIQ), creator of five9-mcp, the same zero-dependency architecture pointed at a second platform. Issues and PRs welcome; the roadmap is the issue tracker.
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.