mcpField
Y
Yuva

Yuva

Remote
useyuva.comCommunication27Updated today

About Yuva

Yuva is an open-source customer messaging platform for teams managing support across multiple products. It combines shared inboxes with publishing, mailing, and analytics, and can be hosted on the team's own servers.

Key features

  1. 01Shared inboxes for email and chat
  2. 02Team mailboxes with search and drafts
  3. 03Mailing lists and campaign reports
  4. 04Help center and status publishing
  5. 05Conversation and website analytics
  6. 06REST API, webhooks, and MCP server

Use cases

  1. 01Triage support conversations across products
  2. 02Draft and review customer replies
  3. 03Publish help and status pages
  4. 04Analyze response times and support trends

MCP support

Yuva includes an MCP server with OAuth for AI assistants such as Claude, Cursor, and VS Code. It can expose customer messaging capabilities to connected assistants.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://{yuva_host}/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Packages
mcpbyuva.mcpb0.4.0

Tools

Details

From the project README
Yuva logo

Yuva

One inbox for every product you run.

Website · Docs · API · Releases · Roadmap · X

Release CI License: AGPL-3.0 SDKs: MIT

Yuva is open-source customer messaging for teams that run several products. Support e-mail, live chat, in-app messages and feedback from every product land in one shared inbox, next to the newsletters, help center, status pages and analytics you send and publish from the same place.

You host it yourself: one Docker image and Postgres.

The Yuva panel: a sidebar with inboxes and views, the conversation list, and an open thread with a teammate's note and a bot's draft reply waiting for review.

Warning

Beta. The API can still change before 1.0, so read the release notes before you upgrade. Database migrations run when the server starts: back up first. The code has not had an independent security audit.

Features

ConversationsOne inbox per product with its own branding, language, hours and team. E-mail, live chat, in-app messages and app feedback in one thread, with notes, labels, mentions and bot drafts a member reviews.
Team mailShared addresses such as sales@ and people's own mailboxes in the same inbox, with drafts, send later, filters and search operators. Personal mailboxes are private, and you can delegate them. Yuva is the mailbox; it is not an IMAP server for mail apps.
ReachMailing lists with consent records and imports, e-mail designs from templates, broadcasts with approval and reports, and a deliverability check.
PublishHelp center, documentation pages, status pages on your own domain, a roadmap board and What's new.
InsightsReports on opened and closed conversations, response times, SLA and ratings, cookieless analytics for websites and apps, and a support report that ranks pages by the conversations they cause.
Move inMigrations of history and live mail from Gmail, Microsoft 365, Yandex, iCloud, Zoho, any IMAP server, Zendesk, Help Scout and Front. The old mailbox is only read.
Build on itA REST API with scoped keys, signed webhooks, identity tokens, headless mode, and an MCP server with OAuth for Claude, Cursor, VS Code and other assistants.

Built with Go, Postgres and a React panel, shipped as one Docker image.

A sent broadcast's report in the Yuva panel: sent, clicks, unsubscribes and clicks per link. The analytics dashboard in the Yuva panel.
The live chat widget on a website.

Quick start

  1. Write compose.yaml and .env from the install guide.

  2. Start it:

    docker compose up -d
    
  3. Create the first owner:

    docker compose exec yuva /yuva bootstrap --email [email protected] --workspace "Example"
    
  4. Put it behind HTTPS and sign in with the code that arrives by e-mail.

On Kubernetes, install the Helm chart oci://ghcr.io/productdevbook/charts/yuva; see Kubernetes with Helm.

SDKs and clients

ForInstallGuide
Web: <yuva-chat>, headless clientnpm install useyuvaWeb widget, Headless
React and shadcn/uinpx shadcn@latest add https://useyuva.com/r/yuva-discussions.jsonshadcn registry
iOS and macOSSwift package https://github.com/productdevbook/yuva.git, product YuvaKitMobile SDKs
Androidimplementation("com.github.productdevbook:yuva:v0.3.1")Mobile SDKs
Gogo get github.com/productdevbook/yuva/sdk/goGo SDK
AI assistants/mcp on your server, or yuva.mcpb from a releaseMCP

Working projects are in examples.

Privacy and security

  • You host it: conversations, contacts and files stay on your servers.
  • Analytics set no cookies and keep no cross-site identifiers; see analytics privacy.
  • Every subscription records who consented, when and where, and lists export with that record.
  • For assistants, customer text is marked as untrusted data, and their replies are drafts unless you allow sending.
  • Report vulnerabilities privately as described in SECURITY.md.

Documentation

Guides, the API reference and examples are at useyuva.com/docs, and in docs/ (also in Turkish and German). How it is built: architecture. What is next: roadmap. What changed: GitHub Releases.

Community and contributing

Read CONTRIBUTING.md first; contributions are made under the CLA. For a local server with Postgres and Mailpit, which catches every mail it sends, run docker compose -f deploy/compose.yaml up -d. Ask questions and report bugs in issues, and follow @useyuva for news.

License

The server and panel are AGPL-3.0. The SDKs (sdk/) and the API contract (openapi/) are MIT, so you can ship them in closed-source apps. Details in LICENSING.md; name and logo in TRADEMARK.md.

“Yuva” is Turkish for nest, or home.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.