Introducing the Transifex MCP Server: Your AI Tools Just Got a Direct Line to Transifex

What's new?
, , , ,
1

Ever wish you could just tell your AI assistant what you need from Transifex instead of tabbing over, clicking through menus, and copy-pasting text back and forth? Now you can.

The Transifex MCP Server connects your favorite AI tools (Cursor, Claude Code, Codex, Windsurf, and more) directly to your Transifex account. Talk to your AI in plain English, and let it handle the rest.

  • “Show me which projects need translating.”
  • “Create a new AI translation task.”
  • “Assign me to this review task.”

You can simply use natural language, no dashboards or clicking around.

What can your AI agent actually do?

Once connected, your AI agent can:

  • Find your work — see every organization and project you have access to
  • Check status — pull project details, target languages, and completion percentages
  • Prioritize tasks — compare projects across your org to spot what has the most untranslated words, so you always know what to tackle first
  • Manage tasks — view, create, or update translation tasks, including setting up automated AI translation
  • Assign yourself — add or remove yourself from human-translation task steps

Getting started

You’ll need:

  1. A Transifex account with access to an organization
  2. A Transifex API token (generate one by following our guide)
  3. An MCP-compatible AI tool, like Claude Code, Cursor, or Codex

Then point your tool at:

https://mcp.transifex.ai/mcp

Cursor example (.cursor/mcp.json):

{
  "mcpServers": {
    "Transifex": {
      "url": "https://mcp.transifex.ai/mcp",
      "headers": { "Authorization": "Bearer <Your-Transifex-API-Token>" }
    }
  }
}

Codex example (.codex/config.toml):

[mcp_servers.transifex]
transport = "streamable_http"
url = "https://mcp.transifex.ai/mcp"
enabled = true

[mcp_servers.transifex.http_headers]
Authorization = "Bearer <Your-Transifex-API-Token>"

Once it’s set up, test it with something like:

“List the Transifex organizations I can access”

If it comes back with your org list, you’re good to go.

Behind the scenes

The server exposes a full toolkit for your AI agent to work with, from get_organization_overview (great for spotting what needs urgent attention across projects) down to granular tools like get_resource_language_stats and set_my_task_assignment. Whether your AI is scanning for the most under-translated project or spinning up a new automated AI translation task, it’s all just a prompt away.

Try it out

Give your AI tool a new superpower today, connect it to Transifex, and start managing your localization workflow without ever leaving your editor. To find out more, feel free to check our help center.

Questions, feedback, or ideas for what you’d like the MCP server to do next? Drop them in the comments below! :backhand_index_pointing_down: