Ethora MCP Server (Model Context Protocol)
MCP serverAI & modelsThis integration connects your AI to Ethora, a chat and messaging platform. Once added, your AI can work directly in your Ethora chats: managing conversations, running AI agents, and automating messaging tasks.
Use Ethora MCP Server (Model Context Protocol) in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Ethora MCP Server (Model Context Protocol) and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use Ethora MCP Server (Model Context Protocol)
Needs your own Ethora MCP Server (Model Context Protocol) account. You sign in to it and approve access when you connect.
Details
Available today. Use it from your connected AI after setup.
After adding the integration, connect it to your Ethora workspace. Then tell your AI what you want handled in your chats and it will take it from there.
What your AI can do with it
- Manage chats and conversations in Ethora
- Run AI agents inside your Ethora chats
- Answer questions using your own knowledge base
- Automate routine messaging tasks
Install Ethora MCP Server (Model Context Protocol)
The server’s own address, for the clients that take one directly. Or connect Ahel once and every client you use reads it from one address, with the account kept on Ahel rather than in each client’s config.
Claude Code
claude mcp add --transport http --scope user ethora-mcp-server-model-context 'https://mcp.chat.ethora.com/mcp/oauth'Run it once in your project, then open /mcp to approve any sign-in the server asks for.
Claude Desktop
https://mcp.chat.ethora.com/mcp/oauthAdd a custom connector in Settings, paste this address, and approve the sign-in.
Cursor
cursor://anysphere.cursor-deeplink/mcp/install?name=ethora-mcp-server-model-context&config=eyJ1cmwiOiJodHRwczovL21jcC5jaGF0LmV0aG9yYS5jb20vbWNwL29hdXRoIn0=Open the link and Cursor adds the server at that address.
ChatGPT
https://mcp.chat.ethora.com/mcp/oauthIn Settings, enable Developer mode, create an MCP app, and paste this address. Your plan and workspace must allow custom apps.
Codex
codex mcp add ethora-mcp-server-model-context --url 'https://mcp.chat.ethora.com/mcp/oauth'Run it once, then sign in with codex mcp login ethora-mcp-server-model-context if the server asks for an account.
From the project's README
As published by dappros/ethora-mcp-server in README.md.
Add the hosted server in one click. You sign in through your browser; there is nothing to install and no key to paste.
For Claude.ai, ChatGPT, Claude Desktop, LM Studio and anything else that takes a connector URL, add
https://mcp.chat.ethora.com/mcp/oauth and sign in. Running it yourself instead? See
Using with stdio clients.
The MCP server for Ethora, an open-source chat and messaging platform with a built-in AI agent framework. It lets Claude, ChatGPT, Cursor, Claude Code, VS Code and autonomous agents create Ethora apps, chat rooms, users and AI agents, post messages, index RAG sources and produce website chat-widget embeds, all through tool calls.
Part of the Ethora SDK ecosystem. Cross-SDK updates: Release Notes. Package changes: CHANGELOG.md.
- npm: https://www.npmjs.com/package/@ethora/mcp-server
- MCP Registry:
io.github.dappros/ethora-mcp-server(https://registry.modelcontextprotocol.io/) - Ethora API (Swagger): https://api.chat.ethora.com/api-docs/#/
Three ways to use it
| Where it runs | Best for | |
|---|---|---|
| Hosted (Ethora Cloud) | https://mcp.chat.ethora.com/mcp (production; https://mcp.chat-qa.ethora.com/mcp is the QA instance) | Claude.ai, ChatGPT, Claude Code, Cursor and agents that talk to Ethora Cloud with no local install |
| Self-hosted | Ships with the Ethora monoserver deploy; enable services.mcp.enabled in deploy.yml and it is served at mcp.<your domain>/mcp | Dedicated or on-premise Ethora installs; agent traffic never leaves your infrastructure |
| stdio CLI | npx -y @ethora/mcp-server on your machine, configured with env vars | Local development, CI, and clients that launch a command |
The hosted and self-hosted modes are the same server started with ETHORA_MCP_TRANSPORT=http. Every MCP session has private in-memory state: one client's login, selected app or tokens are never visible to another session.
Account memory. Some clients open a new MCP session for every tool call (ChatGPT does), which would lose the selected app between calls. So the server keeps a small snapshot per signed-in account, keyed by the user id in the bearer token: the current app and agent, the appToken the API handed that account for the current app, and the enabled tool groups. A new session that arrives with a token for the same account (OAuth, personal URL or Bearer header) gets it back, after the token has been validated against the API. Memory lives in the process, expires after 24 hours idle, and never holds the user token itself. Two more defaults close the same gap: ethora-app-create selects the new app, and an account that owns exactly one app has it selected automatically on the first app-scoped call.
60-second quickstart
Claude.ai or ChatGPT (custom connector). In the Ethora web app open Account, then the AI Assistants tab, create an API key and copy the personal connector URL it shows (https://mcp.chat.ethora.com/mcp/k/<key>). Paste it as a custom connector. Every conversation is authenticated with no login step. For a listed connector that uses the vendor's OAuth login instead, the URL is https://mcp.chat.ethora.com/mcp/oauth.
Claude Code.
# with a personal connector URL (no headers needed)
claude mcp add --transport http ethora https://mcp.chat.ethora.com/mcp/k/<your API key>
# or the open endpoint plus a Bearer header
claude mcp add --transport http ethora https://mcp.chat.ethora.com/mcp --header "Authorization: Bearer <your API key>"
Cursor, VS Code, and any client that takes a URL and headers.
{
"mcpServers": {
"ethora": {
"url": "https://mcp.chat.ethora.com/mcp",
"headers": { "Authorization": "Bearer <your API key>" }
}
}
}
Autonomous agents with no account yet. Connect to https://mcp.chat.ethora.com/mcp with no credentials and call ethora-user-register with an email, first and last name. It creates the account, logs the session in, and returns a generated password plus an API key and connectorUrl exactly once. Store the key and reconnect later with the Bearer header or the personal URL; nothing else is needed, no browser and no email confirmation.
stdio CLI.
ETHORA_API_URL=https://api.chat.ethora.com/v1 ETHORA_APP_JWT="JWT <your app jwt>" npx -y @ethora/mcp-server
Then ask your agent to call ethora-status, ethora-user-login (or ethora-user-register) and ethora-app-list. Lost at any point, call ethora-help: it reads the current state and returns the recommended next calls.
Entry points and authentication
| Entry point | Who supplies identity | Typical client |
|---|---|---|
/mcp (add ?tools=all to list every tool up front) | Nobody at connect time. Call ethora-user-login or ethora-user-register inside the session, or send Authorization: Bearer <token> on every request (user API key, app token or B2B token; the server picks the auth mode from the token type) | Agents, Claude Code, Cursor, connectors added as "no auth" |
/mcp/k/<api-key> | The key in the path, applied like a Bearer header | Claude.ai and ChatGPT custom connectors, which take a URL but no headers |
/mcp/oauth | An OAuth 2.1 access token obtained through the Ethora authorization server (dynamic client registration, PKCE, access scopes read, write, admin, plus the identity scopes openid and email for directories that require them) | Connector directories (Claude, ChatGPT); vendors run the login flow themselves |
| stdio | Env vars ETHORA_APP_JWT (login/register bootstrap) and optional ETHORA_B2B_TOKEN, or ethora-session-configure at runtime | Local CLI |
Stay in user auth mode on the hosted server (ethora-status shows authMode: user). App-token and B2B modes exist for server integrations; the agents and rooms routes reject app tokens.
API keys
ethora-api-key-create { name?, ttlDays? }mints a key (default 90 days, max 365), shown once together withconnectorUrl.ethora-user-registermints one by default andethora-user-login { createApiKey: true }on request.ethora-api-key-listshows id, name, created and expiry, never the value.ethora-api-key-revoke { id }invalidates it immediately: the next request with that key fails withREFRESH_RECORD_NOT_FOUND.- The same keys are managed in the Ethora web app under Account, AI Assistants, where the personal URL, a Claude Code one-liner and a Cursor config are shown with copy buttons.
- A key acts as the user. Treat the personal URL like a password: do not share screenshots of it, revoke it if it leaks. The server never logs request URLs, and the monoserver nginx template logs method and status only on the MCP host.
OAuth 2.1 (/mcp/oauth)
Set ETHORA_MCP_AUTH_ISSUER to the public URL of the Ethora API that hosts the authorization server (the monoserver deploy sets it). The MCP server then:
- serves RFC 9728 protected-resource metadata at
/.well-known/oauth-protected-resourceand/.well-known/oauth-protected-resource/mcp/oauth, naming the authorization server and the three scopes; - answers unauthenticated requests on
/mcp/oauthwith401andWWW-Authenticate: Bearer resource_metadata="...", which is how clients discover the flow; - validates each token against the API once per session (cached five minutes) and enforces the token's
scopeper tool: read-only tools needread; deleting, revoking and rotating (and revealing a credential) needadmin; everything else, including updates that overwrite fields and are therefore marked destructive, needswrite.search,fetch,ethora-help,ethora-statusandethora-doctorneed no scope. Tokens without a scope claim (API keys) get full access; - hides the identity tools (
ethora-user-login,ethora-user-register,ethora-session-configure,ethora-auth-mode-set,ethora-auth-mode-set,ethora-api-key-create,ethora-api-key-list,ethora-api-key-revoke) because the token already fixes who you are.
The authorization server itself is part of the Ethora backend: <issuer>/.well-known/oauth-authorization-server, /oauth/register, /oauth/authorize (a consent page with sign-in, account creation and Google sign-in), /oauth/token, /oauth/revoke and /oauth/userinfo. Users see and disconnect OAuth grants under Account, AI Assistants, Connected AI apps. When ETHORA_MCP_AUTH_ISSUER is unset, both OAuth routes return 404 and discovery omits them.
Identity scopes. Some directories (ChatGPT) require the minimal OpenID Connect surface on top of OAuth 2.1: the openid and email scopes and a userinfo endpoint that returns sub, email and email_verified. These scopes grant no access to apps or data; they only let the client see who signed in, and the consent page says so in plain words. There are no ID tokens or JWKS, because nothing consumes them.
Email confirmation is optional. Ethora never blocks sign-up or the dashboard on a confirmed address. When a client asks for the identity scopes and the account's address is not yet confirmed, the consent page adds one step: send the confirmation link, continue after confirming, or continue without sharing the address (the identity scopes are dropped from the grant and everything else proceeds). Google sign-ins arrive confirmed and skip the step. The same confirmation can be sent from Account, AI Assistants in the web app.
What agents can do
The end-to-end journey a new user typically asks for, with the tools in order:
ethora-user-register(orethora-user-login) to get an authenticated session and an API key.ethora-app-create { displayName }thenethora-app-select { appId }to make the new app current.ethora-chat-create { title }to create a group room. The result contains the room JID${appId}_${chatId}; every room tool accepts the JID or the barechatId.ethora-agent-create { name, prompt, ... }to create an AI agent persona in that app.ethora-agent-invite { agentIdOrAddress, chatJid }to put the agent in the room. A bot instance is spawned live, no restart needed.ethora-message-send { text, roomJid, waitForReplySec: 45 }to post a message and wait for the agent's answer, returned asreplies.ethora-chat-historyreads the room afterwards.ethora-agent-activate { agentId, chatJid }to make that agent the app's default responder, thenethora-widget-snippet-getfor the<script>tag that puts the AI chat widget on a website.
ethora-help { goal } returns this and the other recipes (user-login, broadcast, sources-ingest, files-upload, bot-manage, chat-test, widget, b2b-bootstrap-ai) with the calls filled in for the current state, and ethora-recipe-run executes them.
First-minute conventions
- Every create tool answers the same way.
ethora-app-create,ethora-chat-create,ethora-agent-createreturn the raw API object pluscreated(kind,id,name, andjidoraddresswhere relevant) andnext, two to four suggested calls with arguments filled in.ethora-app-createalso returnsdashboardUrl, the app in the web dashboard. ethora-help { goal }has a recipe for each headline job:new-app,in-app-chat,multi-agent-room,widget,chat-test, plus the server-integration goals. Each returns the calls in order with arguments to copy.- Who spoke.
ethora-message-sendreplies andethora-chat-historyrows carrysenderNameandsenderKind(human,agent,app), so a multi-agent room reads as "Freud: ..., Jung: ..." rather than instance ids. - Search understands intent.
searchmaps the phrases people use ("add chat to my React app", "several agents talking to each other", "webhook when a message arrives") to the documents that answer them, anddoc:not-availablesays plainly what is not exposed over MCP and where it lives instead. Whole documents are fetchable by their bare id (doc:recipes,doc:chat-component-quickstart,doc:sdk-backend-quickstart,doc:auth-map).
Documentation inside the server
instructionsin the initialize result tell the assistant how identity works on the entry point it connected through: the open endpoint explains login and register, personal-URL and Bearer sessions are told they are already authenticated and must never ask for a password or key, OAuth sessions the same plus how to react toINSUFFICIENT_SCOPE.search { query }andfetch { id }(the ChatGPT connector convention) search an in-memory corpus: the auth map, quickstarts, recipes, a hosted getting-started guide, an API keys guide and one reference entry per tool with its inputs. They work unauthenticated.- Resources
ethora://docs/auth-map,ethora://docs/chat-component/quickstart,ethora://docs/sdk-backend/quickstart,ethora://docs/recipesand promptsethora-auth-map,ethora-vite-quickstart,ethora-nextjs-quickstart,ethora-backend-sdk-quickstart,ethora-recipes,ethora-agents-quickstart.
Tool groups
A session lists the core group only at first: 24 tools covering the whole "sign in, create an app, add rooms and messages, create and activate an agent, give it a knowledge base, get the widget" journey, one variant per operation. The other groups are registered but hidden, which keeps tools/list around 40 KB instead of 120 KB and gives assistants a short list to choose from.
Three ways to get more:
ethora-tools-enable { group }enables a group for the session (or{ group: "all" }); the server sendstools/list_changedand the client refreshes. With no arguments it returns the catalogue with counts.- Calling a hidden tool by name enables its group and runs it, so a name learned from the docs or an earlier session is never refused.
?tools=allon the endpoint URL (hosted) orETHORA_MCP_TOOLS=all(stdio) lists everything up front.
search and fetch describe hidden tools too; every tool doc names its group, and doc:tool-groups is the catalogue. Legacy and async variants carry a first line naming the preferred sibling.
| Group | What it covers | Tools |
|---|---|---|
core (listed by default) | Sign in or register, create an app, add rooms and messages, create and activate an AI agent, give it a knowledge base, get the website widget. Always listed. | ethora-status, ethora-help, ethora-feedback-submit, ethora-tools-enable, search, fetch, ethora-user-register, ethora-user-login, ethora-api-key-create, ethora-app-create, ethora-app-list, ethora-app-select, ethora-app-update, ethora-chat-create, ethora-message-send, ethora-chat-history, ethora-agent-create, ethora-agent-list, ethora-agent-update, ethora-agent-invite, ethora-agent-activate, ethora-source-site-crawl-wait, ethora-source-doc-upload, ethora-widget-snippet-get |
keys | List and revoke API keys, reveal an app's credentials, mint and rotate app tokens. | ethora-api-key-list, ethora-api-key-revoke, ethora-app-credentials-reveal, ethora-app-token-create, ethora-app-token-list, ethora-app-token-revoke, ethora-app-token-rotate |
session | Diagnostics, recipes and switching the session's auth mode (app token, B2B token) for server integrations. | ethora-doctor, ethora-recipe-run, ethora-session-configure, ethora-auth-mode-set, ethora-auth-mode-set, ethora-auth-mode-set |
apps-admin | Delete, export and import whole apps; inspect default rooms. | ethora-app-delete, ethora-app-export, ethora-app-import, ethora-app-rooms-list, ethora-app-rooms-list |
rooms | Delete rooms, broadcast to many rooms, search messages, read message context and unread counts. | ethora-chat-delete, ethora-broadcast-send, ethora-broadcast-job-start, ethora-broadcast-job-wait, ethora-message-search, ethora-message-context, ethora-chat-unread-counts |
agents-admin | Inspect, clone (with or without its knowledge), delete, export and import agents; try an agent without a room; check and rebuild its knowledge index; edit its Memory and visibility. | ethora-agent-get, ethora-agent-clone, ethora-agent-delete, ethora-agent-export, ethora-agent-import, ethora-agent-visibility-set, ethora-agent-soul-set, ethora-agent-soul-append, ethora-agent-try, ethora-agent-knowledge-get, ethora-agent-knowledge-rebuild |
sources | Knowledge-base maintenance: async crawl and reindex jobs, stop a crawl, list and tag sites and documents, delete URLs and documents. | ethora-source-site-crawl, ethora-source-site-crawl-cancel, ethora-source-site-reindex, ethora-source-site-reindex-wait, ethora-source-site-list, ethora-source-site-tags-update, ethora-source-site-url-delete, ethora-source-site-url-delete-batch, ethora-source-doc-list, ethora-source-doc-tags-update, ethora-source-doc-delete |
widget | Read and change the website widget's saved look (colours, fonts, size, launcher, teaser, texts); live embeds pick it up without a new snippet. | ethora-widget-appearance-get, ethora-widget-appearance-set |
users-files | Batch-create users and upload, fetch or delete files. | ethora-user-batch-create, ethora-user-batch-job-start, ethora-user-batch-job-wait, ethora-file-upload, ethora-file-get, ethora-file-delete |
legacy-bot | The per-app bot of apps created in the dashboard before the agents framework, and the pre-v2 document tools. Prefer the agents and sources tools for anything new. | ethora-bot-get, ethora-bot-update, ethora-bot-enable, ethora-bot-disable, ethora-bot-widget-get, ethora-bot-history, ethora-bot-message-send, ethora-bot-instance-list, ethora-bot-instance-status-set, ethora-bot-instance-diagnose, ethora-bot-instance-leave, ethora-bot-instance-test, ethora-bot-enable-b2b, ethora-source-doc-upload-legacy, ethora-source-doc-delete-legacy |
b2b | Server-to-server provisioning with a B2B token, plus code and config generators for integrations. | ethora-b2b-app-create, ethora-b2b-app-provision, ethora-b2b-app-bootstrap-ai, ethora-b2b-runbook-generate, ethora-env-examples-generate, ethora-chat-component-app-generate, ethora.b2b.auth.use, ethora.b2b.app.create, ethora.b2b.bot.enable, ethora.b2b.broadcast.wait, ethora.b2b.app.bootstrap-ai |
wallet | Wallet balance and ERC-20 transfer. Local (stdio) only; never offered on the hosted server. | ethora-wallet-balance-get, ethora-wallet-erc20-transfer |
App deletion and bulk-delete tools are only registered when ETHORA_MCP_ENABLE_DANGEROUS_TOOLS=true (the monoserver deploy sets it; the stdio default is off). Alias tools (ethora.b2b.*, ethora-bot-message-send, ethora-bot-history) are off by default (ETHORA_MCP_ENABLE_ALIASES=true to expose them); the canonical tools cover the same ground.
Website widget
ethora-widget-snippet-get returns the tag for the embeddable AI chat widget:
<script id="chat-content-assistant" src="https://widget.<your domain>/assistant.js"
data-app-id="<appId>" data-api-base="https://api.<your domain>" defer></script>
The widget's look (colours, fonts, size, launcher, teaser bubble, texts) is saved on the app: ethora-widget-appearance-set { appearance: { "data-primary-color": "#7c3aed" } } (the widget group), the same setting as App settings > AI Widget in the admin panel. Every embed loads it when it opens, so later changes reach live sites without a new snippet. A data-* attribute written on the tag wins over the saved value on that site.
The widget answers with the app's active bot (defaultBotInstanceId). On an app created through the API run ethora-agent-create, ethora-chat-create, ethora-agent-invite and ethora-agent-activate { agentId, chatJid } first; the tool lists these prerequisites and ethora-help { goal: "widget" } walks through them. Activation runs in user auth by setting the app's default bot instance (what the admin AI Widget dropdown does). Until a bot is active, the widget's session endpoint answers AI_BOT_NOT_CONFIGURED. ethora-chat-component-app-generate produces a React App.tsx for @ethora/chat-component instead.
Configuration
Env vars (stdio and hosted)
| Variable | Meaning |
|---|---|
ETHORA_API_URL | Full API URL, e.g. https://api.chat.ethora.com/v1 (default). On a hosted server it is fixed for all sessions |
ETHORA_BASE_URL | Host-only alternative to ETHORA_API_URL; /v1 is appended |
ETHORA_APP_JWT | App JWT used only by login and register (ETHORA_APP_TOKEN is a legacy alias) |
ETHORA_APP_DOMAIN_NAME | Base app domainName; when ETHORA_APP_JWT is empty the server fetches the app JWT from GET /v1/apps/get-config?domainName=... at startup |
ETHORA_B2B_TOKEN | B2B server token for x-custom-token tenant-actor routes |
ETHORA_MCP_TOOLS | all lists every tool from the start instead of the core group (stdio; hosted uses ?tools=all on the URL) |
ETHORA_MCP_ENABLE_DANGEROUS_TOOLS | true registers app deletion, wallet transfer and bulk-delete tools (default off) |
ETHORA_MCP_OPENAI_APPS_CHALLENGE | Hosted only. Token issued by the OpenAI apps portal for domain verification; served verbatim at /.well-known/openai-apps-challenge (404 when unset) |
ETHORA_MCP_ENABLE_ALIASES | true exposes the dot-namespaced alias tools (default off) |
Hosted mode only
Shortened here. Read the whole README on GitHub.
Signals
- GitHub stars
- 6
- Forks
- 1
- Last commit
- Sep 2026
- Weekly_downloads
- 824 weekly_downloads
Advanced
- Delivery
- ethora-mcp-server MCP server → your Ahel connector (mcp.ahel.ai) → your AI.
- Item type
- mcp-server
- Key
io-github-dappros-ethora-mcp-server- Source
- github.com/dappros/ethora-mcp-server
- Hosted endpoint
https://mcp.chat.ethora.com/mcp/oauth
github.com/dappros/ethora-mcp-server
More in AI & models
MCP server · upstash
More in AI & modelsAdButler
MCP server · adbutler
More in AI & modelsatom-mcp-server
MCP server · a7om-ai
More in AI & modelsottasia
MCP server · aiweather-anurag
More in AI & modelsunphurl
MCP server · 123ergo
More in AI & modelsmicrosoft-learn-mcp
MCP server · microsoftdocs
More in AI & models