Log inSubmit server
Front MCP logo

Front MCP

Connects MCP-compatible clients to Front for conversation, contact, messaging, and inbox administration.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Front MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "front-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@houst-com/front-mcp"
      ],
      "env": {
        "FRONT_API_TOKEN": "YOUR_VALUE_HERE",
        "FRONT_MCP_AUTH_METHOD": "YOUR_VALUE_HERE",
        "FRONT_MCP_OAUTH_SECRET": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

npx -y @houst-com/front-mcp

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 12d ago.

Overview

The Front MCP MCP server exposes Front API operations through MCP tools for conversations, contacts, messages, tags, and inbox administration. It runs over stdio and authenticates with a Front API token or an OAuth application. Reach for it when an agent needs to search threads, manage customer records, draft or send replies, assign work, or maintain Front configuration. Write operations use confirmation checks, while destructive operations are denied by default.

Use cases

  • Search customer conversations by topic
  • Manage contacts and their conversation history
  • Draft or send replies in Front
  • Assign and tag inbox conversations
  • Maintain Front teams, inboxes, and knowledge bases

Key features

  • Front conversation search and management
  • Contact, message, draft, and comment operations
  • Inbox, teammate, team, and channel administration
  • Tags, templates, signatures, and knowledge-base management
  • OAuth and API-token authentication
  • Policy controls for write and destructive actions

Tools26

What Front MCP lets your agent do. Expand a tool for its parameters.

Self-reported~529 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

10/10 checks healthy over the last 49d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 18h ago via OSV.dev · @houst-com/front-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads498Package downloads in the last 30 days.
Last commit1d agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Tools exposed26Callable tools this server registers over MCP.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Front MCP works

What Front MCP MCP server does

The Front MCP MCP server connects an MCP-compatible client to one Front account and makes Front API resources available as grouped tools. It covers conversation search and updates, contact records, messages, drafts, comments, tags, inboxes, teammates, teams, channels, and account relationships.

The available groups also include analytics exports and reports, custom fields, contact groups and lists, knowledge bases, links, rules, shifts, signatures, message templates, events, and token identity. Actions range from reading records to creating, updating, assigning, replying, importing, and deleting them.

This is suited to operational workflows where an agent needs to inspect customer conversations and take controlled actions in Front. Example tasks include finding billing-related threads, reading a conversation's messages, locating a contact by email, tagging and assigning a conversation, or preparing a draft reply.

How it works

The package runs as a local stdio MCP process. An MCP client starts it with npx, and the process communicates with that client through the MCP protocol while making outbound HTTPS requests to the Front API. Each server instance is limited to one Front account, and requests are not cached; current data is retrieved from Front for each request.

Authentication supports a Front API token or OAuth. OAuth credentials can be stored through the authentication command, which opens a browser-based flow. Stored tokens are encrypted locally with AES-256-GCM and use owner-only file permissions.

Actions are categorized as read, write, or destructive. Reads are allowed by default. Writes normally require a confirmation step, and destructive operations are denied unless policy configuration changes that decision. A policy file can set defaults and action-specific overrides.

Setup and configuration

For token authentication, provide FRONT_API_TOKEN in the MCP client's environment and start the package with npx -y @houst-com/front-mcp. The token is created in Front under Settings, Developers, and API tokens.

For OAuth, create a Front OAuth app, configure the documented localhost callback, select the required resource permissions, and set FRONT_MCP_AUTH_METHOD=oauth plus FRONT_MCP_OAUTH_SECRET. Run npx @houst-com/front-mcp auth to complete sign-in. FRONT_MCP_LOG_LEVEL controls logging, and FRONT_MCP_POLICY_FILE points to a custom policy JSON file.

Front permissions must match the requested actions: Read covers list, get, and search operations; Write covers creation, updates, assignment, additions, merging, and replies; Delete covers deletion and removal; Send covers message creation and replies.

Tools and capabilities

  • Search, read, create, update, assign, and remove Front conversations.
  • List, maintain, merge, and associate contacts with conversations.
  • Create messages, replies, drafts, comments, notes, tags, and links.
  • Manage inbox access, channels, teammates, teams, shifts, signatures, and rules.
  • Maintain knowledge-base articles and categories, message templates, and custom fields.
  • Produce analytics exports and reports.

Limitations and notes

The Front MCP MCP server currently supports stdio only; HTTP and SSE transports are not available. It does not process inbound webhooks, upload or download attachments, or provide application-channel-specific message synchronization endpoints. Every request reaches the Front API directly.

The server writes no secrets to standard output, reserves that stream for MCP traffic, and supports configurable output and log redaction. Destructive actions remain denied by default unless an explicit policy changes the behavior.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Front MCP

Front MCP is an MCP server that connects MCP-compatible clients to Front. Its main tools cover conversations, contacts, messages, drafts, comments, tags, inboxes, teammates, teams, channels, knowledge bases, templates, analytics, and related Front resources, including read, create, update, assignment, reply, and deletion actions.

Install it by configuring an MCP client to run `npx -y @houst-com/front-mcp`. Add `FRONT_API_TOKEN` to the server environment, then start the client. The README provides a Claude Code configuration example using this command and environment variable.

Use OAuth by creating a Front OAuth app, setting its localhost callback, selecting resource permissions, and configuring `FRONT_MCP_AUTH_METHOD=oauth` with `FRONT_MCP_OAUTH_SECRET`. Then run `npx @houst-com/front-mcp auth` to complete the browser-based authentication flow.

Yes, Front MCP works with Claude Code through its stdio configuration. Add the package command and either `FRONT_API_TOKEN` or the OAuth environment variables to `~/.claude/settings.json`, then restart Claude Code to expose the Front tools.

No, the current version supports stdio transport only. HTTP and SSE transports are listed as unavailable in version one, so the MCP client must launch the package as a local process.

Related MCP servers

All in Support & Service Management All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewFront MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/front-mcp?style=directory)](https://allmcps.com/mcp/front-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/front-mcp"><img src="https://allmcps.com/api/badge/front-mcp?style=directory" alt="Front MCP on AllMCPs" /></a>