MCP Server for Chat, AI Agents & Messaging

No need to switch between dashboards or manually wire up APIs — tell your AI what you want to do. Ethora’s Model Context Protocol (MCP) Server connects AI IDEs and agents (Cursor, VS Code MCP, Claude Desktop, Windsurf) to Ethora tools: Apps, Chats, Broadcasts, Sources (RAG), Files, Bots, Wallets.

Ethora MCP Server for Cursor, VS Code MCP, Claude Desktop

Why MCP matters

Every agent, model, and framework speaks a different language. MCP is a standard protocol that lets them share context and use tools safely and consistently.

With Ethora MCP Server, your AI can:

  • call Ethora REST APIs through approved tools (instead of ad-hoc code)
  • load integration docs into context (resources)
  • use repeatable "how to integrate" prompts (prompts)
  • generate copy/paste snippets (generators)

Get started (2 minutes)

Get Ethora MCP Server running in seconds. Compatible with Cursor, VS Code MCP, Claude Desktop, and Windsurf.

1

Run the server

npx -y @ethora/mcp-server
2

Configure it (env vars)

Set these in your MCP host environment (or configure at runtime using the ethora-configure tool):

# Required: Ethora API endpoint ETHORA_API_URL=https://api.ethoradev.com/v1 # Choose one authentication method: ETHORA_APP_JWT="JWT <APP_JWT_FOR_LOGIN_REGISTER>" ETHORA_B2B_TOKEN="JWT <B2B_SERVER_TOKEN>"
3

Connect from your IDE

Add "Ethora MCP Server" in your MCP client settings and point it to the local command:

npx -y @ethora/mcp-server

What you get (actual capabilities)

Apps (B2B automation)

Create, update, list, delete apps
Select a "current app" context for subsequent tools

Chats & messaging

Broadcast a message to many rooms via a background job
Poll job status and per-room results

Sources ingestion (RAG automation)

Crawl a website URL for indexing
Reindex and delete previously ingested URLs
Upload documents for ingestion, delete ingested documents
(Includes app-token variants so integrations can run without user credentials)

Files

Upload, list/get, delete files (user-auth flows)

Bot / Agent management

Get/update bot configuration; enable/disable bot for an app (app-token flows)

Wallet tools (Web3)

Get balances; ERC-20 transfers

Auth map: appJwt vs appToken vs b2bToken

Ethora supports multiple auth modes depending on who is calling.

App JWT (appJwt)

Use for: user login/register bootstrap
Where it lives: MCP host env (ETHORA_APP_JWT) or ethora-configure
Typical caller: your frontend + backend during initial user auth

App Token (appToken)

Use for: app-scoped B2B operations (broadcast / sources ingest / bot management)
Where it lives: per-app secret in your infra / admin config
Typical caller: partner integrations automating a single app

B2B Token (b2bToken)

Use for: server-to-server automation (create apps, bootstrap environments)
Sent as: x-custom-token header (JWT with type=server)
Where it lives: secure backend/MCP host secret store (ETHORA_B2B_TOKEN)

Rule of thumb

  • Need to log in a user → App JWT
  • Need to automate one app (broadcast/sources/bot) → App Token
  • Need to create apps / bootstrap partner flows → B2B Token → then App Token
Field appJwt appToken b2bToken
Purpose User-level operations Application-level operations Organization-level operations
Where Used User-specific actions, personal assistants Backend services, automation, AI agents Enterprise automation, multi-tenant systems
Where Stored User session or secure storage Environment variables, secrets manager Enterprise secrets vault, CI/CD
Best Practices Rotate regularly, use HTTPS only Store in .env (never commit), use service accounts Use least-privilege tokens, audit access
Environment Variable ETHORA_APP_JWT ETHORA_APP_JWT ETHORA_B2B_TOKEN

Prompts & Resources (built-in integration docs)

Ethora MCP Server includes developer-facing prompts and loadable resources that mirror how people actually integrate Ethora.

Resources (load docs into the AI context)

ethora://docs/auth-mapComplete authentication guide and token comparison
ethora://docs/chat-component/quickstartQuickstart guide for Ethora Chat Component integration
ethora://docs/sdk-backend/quickstartBackend SDK integration guide and examples
ethora://docs/recipesCommon integration patterns and code recipes

Prompts (reusable integration prompts)

ethora-auth-mapUnderstand authentication modes and token usage
ethora-vite-quickstartGenerate Vite + React integration code
ethora-nextjs-quickstartGenerate Next.js integration code
ethora-backend-sdk-quickstartGenerate backend SDK integration code
ethora-recipesAccess common integration patterns and examples

Related Documentation

Explore additional resources to enhance your integration

Generators (no shell, no file writes)

These tools output copy/paste artifacts directly in your MCP client.

ethora-generate-chat-component-app-tsx— ready-to-paste App.tsx snippet
ethora-generate-env-examples— .env.example templates (frontend / backend SDK / MCP)
ethora-generate-b2b-bootstrap-runbook— ordered runbook of MCP tool calls

ethora-generate-chat-component-app-tsx

Generates a complete React TypeScript component for integrating Ethora Chat Component. Produces ready-to-use JSX/TSX code that you can copy into your application.

Output: TypeScript React component with proper imports, hooks, and configuration.

ethora-generate-env-examples

Generates example environment variable configurations for different authentication modes and use cases. Helps you set up .env files correctly.

Output: Formatted environment variable examples with comments and best practices.

ethora-generate-b2b-bootstrap-runbook

Generates a step-by-step runbook for bootstrapping a B2B integration, including authentication setup, app creation, and initial configuration.

Output: Markdown-formatted runbook with commands, code examples, and verification steps.

Note: Generator tools produce code and documentation only. They do not execute commands or modify your filesystem. You control when and how to use the generated code.

Common recipes

1

Broadcast to rooms (app-token)

  1. Select app context (sets appId and appToken)
  2. Switch to app-token auth
  3. Enqueue broadcast job
  4. Poll job status until complete
2

Ingest a website (RAG) (app-token)

  1. Select app context
  2. Switch to app-token auth
  3. Run site crawl, then reindex later if needed
3

Full B2B bootstrap (create app → ingest → enable bot)

  1. Configure API URL + B2B token
  2. Switch to B2B auth
  3. Run bootstrap tool (creates app + indexes sources + enables bot)
  4. Switch to app-token auth for follow-up bot tuning
1

Broadcast Message

App-Token Auth • Send a broadcast message to a chat room

# 1. Set environment variables export ETHORA_API_URL=https://api.ethora.com export ETHORA_APP_JWT=your_app_token_here # 2. Start MCP server npx -y @ethora/mcp-server # 3. In your AI agent (Cursor/Claude Desktop), use: # "Send a broadcast message to room [roomId] with text 'Hello, world!'" # Or via tool call: { "tool": "ethora_chat_broadcast", "arguments": { "roomId": "your-room-id", "message": "Hello, world!", "messageType": "text" } }
2

Sources Ingestion / RAG

App-Token Auth • Ingest documents and URLs for RAG-powered AI applications

# 1. Configure app-token auth export ETHORA_API_URL=https://api.ethora.com export ETHORA_APP_JWT=your_app_token_here # 2. Start MCP server npx -y @ethora/mcp-server # 3. Crawl a URL for indexing # "Crawl and index https://docs.example.com/api-reference" # 4. Upload a document # "Upload document 'user-manual.pdf' to source collection 'docs'" # 5. Reindex existing sources # "Reindex all sources in collection 'docs'" # Tool calls example: { "tool": "ethora_sources_crawl", "arguments": { "url": "https://docs.example.com/api-reference", "collectionId": "docs" } }
3

Full B2B Bootstrap

B2B → App-Token • Complete workflow for bootstrapping a B2B integration

# Step 1: B2B Authentication Setup export ETHORA_API_URL=https://api.ethora.com export ETHORA_B2B_TOKEN=your_b2b_token_here # Step 2: Start MCP server with B2B auth npx -y @ethora/mcp-server # Step 3: Create application via MCP # "Create a new Ethora application named 'My B2B App'" # Step 4: Get app token from created app # The app creation response includes app credentials # Step 5: Switch to app-token auth for app operations export ETHORA_APP_JWT=app_token_from_step_4 export ETHORA_B2B_TOKEN="" # Clear B2B token # Step 6: Restart MCP server with app-token npx -y @ethora/mcp-server # Step 7: Configure app features # "Configure chat rooms, sources, and bot settings for app [appId]" # Complete workflow via tool calls: # 1. Create app (B2B auth) { "tool": "ethora_apps_create", "arguments": { "name": "My B2B App", "description": "Enterprise automation app" } } # 2. Configure app (app-token auth) { "tool": "ethora_chat_create_room", "arguments": { "appId": "created-app-id", "name": "Support Channel" } }

What is Model Context Protocol (MCP)?

Every agent, model, and framework speaks a different language. MCP is a standardized protocol that enables them to share context and collaborate effectively. Ethora MCP Server implements the MCP specification, allowing AI IDEs like Cursor, VS Code MCP, Claude Desktop, and Windsurf to access Ethora platform capabilities through a unified interface.

AI IDE connected to Ethora platform through the MCP server

Why MCP Matters

Lack of a common framework leads to one-off integrations, which are difficult to maintain and reuse. MCP eliminates this friction and provides developers with:

  • Standardization across AI models and platforms
  • Reusable connectors for multi-agent systems
  • Secure tool usage and context management
  • Developer-friendly AI interaction layer

Connect Any AI Agent or System

Ethora MCP Server works with the most popular AI-assisted development tools. It doesn’t matter what LLM model you use for coding—Ethora integrates seamlessly.

Supported Hosts Compatible AI Agents and Models
CursorClaude (Anthropic), GPT-4, custom models
VS Code MCPAny model supporting MCP protocol
Claude DesktopClaude 3 (Opus, Sonnet, Haiku)
Windsurf IDEIntegrated AI models
Custom AgentsAny AI agent that communicates via MCP

Real-World Use Cases

Multi-agent AI Systems

Use standardized tools and prompts to coordinate agents in real-time. Build complex workflows where multiple AI agents collaborate on tasks.

B2B Automation

Automate processes for digital assets, documents, and chat. Integrate Ethora capabilities into enterprise workflows and systems.

RAG Ingestion & Knowledge Bases

Ingest documentation, support articles, and knowledge bases for AI-powered search and Q&A systems. Build context-aware assistants.

AI-IDE Development

Leverage Ethora’s extensive functionality by letting AI create your app through "vibe coding." Build faster with AI-assisted development.

Enterprise-Grade Security

We understand that keeping your data safe is crucial in the modern interconnected world. That’s why Ethora MCP Server uses the best security practices.

End-to-end encryption

Data is encrypted when transferred and stored.

Access control and authentication

Multiple authentication modes prevent unauthorized access and misuse.

Compliance

Alignment with AI governance and industry standards.

Data privacy protection

Across all deployments and authentication modes.

How to Get Started with MCP Server

Whether you’re using Cursor, VS Code MCP, Claude Desktop, or another AI-assisted IDE, the process is almost the same: run the server locally and point your environment to it.

  1. Install the MCP Server: npx -y @ethora/mcp-server
  2. Configure environment variables (ETHORA_API_URL, ETHORA_APP_JWT or ETHORA_B2B_TOKEN)
  3. Add Ethora as a new MCP server in your IDE’s settings
  4. Start building with access to Ethora’s features
Pricing

Save Months of Work with Ethora

Build any chat use case into your product: in hours.

Free

$0 / mo
free forever
1,000 MAU
100 Concurrent Connections
Sign Up
All Chat & AI features, including:
  • Community Support
  • 30 Days of Free Support
  • No Credit Card Required
Recommended

Enterprise

$599+ / mo
billed monthly
Unlimited MAU
Unlimited Connections
All Business, plus:
  • 24/7 Phone Support
  • Self-hosted AI
  • Custom Configuration
  • Dedicated / On-prem hosting

What our clients say about our services.

Taras and the Ethora team are a reliable partner in messaging, tokenization and AI. Great communication and products.

Joe Sticca
Joe Sticca
Digital Product & Technology Leader, Trust Industries

I have worked with Ethora and team since 2018. They have been absolutely amazing and life-saving across multiple of my portfolio projects.

Adam Palmer
Adam Palmer
Co-founder, SIAD Ventures

We partnered with Dappros/Ethora on an EU project and they successfully delivered the technology component. Highly recommended.

Peter Fearon
Peter Fearon
CEO, Remade Group

Ethora has successfully helped us ideate and deliver multiple digital transformation projects in Qatar and the MENA region. A reliable and knowledgable partner.

Tariq Gulrez
Tariq Gulrez
Ministry of Communications & IT

Ethora technology and expertise were instrumental in delivering Atom Connect, our product for streamlined communication in the workers’ comp market.

Carol Valentic
Carol Valentic
CEO, Candollar

Ethora helped us deliver our AI agent solution, letting subscribers access context-specific insights from our vast library of expert content. Their LLM and chat-bot expertise is an excellent match for our AI strategy.

Bret Gregory
Bret Gregory
CEO, DrTalks

Ethora has been an exceptional strategic development partner for Preshent. Not just a vendor, but a core contributor to the future of our platform and company.

John Richardson
John Richardson
Founder & CEO, Preshent Corporation

Ethora helped us ideate and deliver multiple digital transformation projects across Qatar and the MENA region. A reliable, knowledgable partner.

Howaida Nadim
Howaida Nadim
CEO, Foresight Communications

Our Case Studies

DrTalks
DrTalks Logo

Dr. Talks is one of the most prominent organizers of online healthcare conferences and summits.

Ethora enabled them to create a context-aware AI-powered chatbot to help users navigate thousands of pages of medical content, correctly referencing pages, authors, videos and timestamps. 100% secure and HIPAA-compliant.

This assistant enhances user experience by simplifying interaction with the platform and reducing search time.

  • 1000+ summits indexed by AI bot
  • Instant user engagement
  • Vector embeddings enable intelligent domain expert AI responses
Read more
Atom
Atom Logo

Atom Advantage is a provider of innovative AI powered workers compensation and health insurance solutions in the United States.

Ethora engine has enabled Atom Connect product used for communication between injured workers, nurses, and caseworkers. AI-powered, with documents exchange, and compliant with all regulations such as HIPAA and SOC2.

This solution reduces manual labour for caseworkers and improves patients experience.

  • Integration into caseworkers web portal
  • White-labelled mobile app for chat and health documents wallet for injured workers
  • Instant messaging between caseworkers and injured workers
Read more

Implementation options

Local (recommended for developers)

Run via npx and connect from Cursor/VS Code/Claude Desktop

Self-hosted

Run inside your infrastructure for centralized governance and shared access controls

Docker/Kubernetes

For enterprise environments and scaling

Security notes

  • Never commit tokens to git.
  • Prefer env vars / secret managers / your IDE’s secure secret store.
  • For production integrations, prefer App Token / B2B Token over sharing user credentials.
Get started

Start building with Ethora MCP

Run it locally and connect from your AI IDE:
npx -y @ethora/mcp-server

Free tier available Enterprise SLA No vendor lock-in

FAQ

Do I need to know the Ethora API to use the MCP server?

No. Your AI can work through the MCP tools and built-in prompts/resources. You can still use the API directly if you want.

Can I use it without user credentials?

Yes. Many automation features support app-token and B2B flows.

Which IDEs are supported?

Any MCP host, including Cursor, VS Code MCP, Claude Desktop, and Windsurf.

What are the main use cases for Model Context Protocol Server?

The Ethora MCP Server enables AI agents to act as intelligent platform assistants that can:

  • Manage community interactions
  • Provide customer support through real-time messaging and smart responses
  • Facilitate transactions using Web3 tools and digital assets
  • Moderate content
  • Employ gamification and social features to improve engagement and UX
  • "Vibe-coding" (Automate development tasks via AI-assisted IDEs)
  • B2B automation and enterprise workflows
  • RAG ingestion and knowledge base management