Skip to content
Back to search

lara-mcp

A Model Context Protocol (MCP) server for Lara Translate, enabling professional translation capabilities with support for language detection, context-aware translations, translation memories, and…

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/translated/lara-mcp with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/translated/lara-mcp'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/translated--lara-mcp/badge.svg)](/listings/translated--lara-mcp)

HTML

<a href="/listings/translated--lara-mcp"><img src="/api/listings/translated--lara-mcp/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Pick your client below — no API keys needed, just log in through your browser.

Claude Desktop

1. Go to **Settings** > **Connectors** 2. Click **Add Custom Connector** 3. Enter the name: `Lara` 4. Enter the URL: `https://mcp-v2.laratranslate.com/v1` 5. Click **Add**, then click **Connect** 6. Log in with your Lara Translate credentials in the browser

Done — Lara Translate is now available in your conversations.

Cursor

Once Lara Translate is listed in the official Cursor plugin marketplace, install it from inside Cursor:

1. Open the plugin browser and search for **Lara Translate**. 2. Click **Install**. 3. The first time you use a Lara tool, your browser will open to authenticate.

In the meantime, or for manual installation, see the [Client Setup Guide](docs/client-setup.md#cursor).

Claude Code

Once Lara Translate is listed in the official Claude Code plugin marketplace, install it from inside Claude Code:

/plugin

Search for **Lara Translate** and install. The first time you use a Lara tool, your browser will open to authenticate.

In the meantime, or for manual installation, see the [Client Setup Guide](docs/client-setup.md#claude-code).

Other Clients

For step-by-step OAuth setup on **VS Code (GitHub Copilot)**, **Windsurf**, **Cline**, **Continue**, and more, see the **[Client Setup Guide](docs/client-setup.md)**.

If your client isn't listed, the general approach is to add the server URL (`https://mcp-v2.laratranslate.com/v1`) to your MCP config — the client will handle OAuth authentication automatically.

> For a complete list of MCP-compatible clients, visit the [official MCP clients page](https://modelcontextprotocol.io/clients).

Verify It Works

After setup, test with a simple prompt:

Translate with Lara "Hello world" to Spanish

Your client should invoke Lara Translate and return the translation.

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 334 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. # Lara Translate MCP Server
  2. A Model Context Protocol (MCP) server for [Lara Translate](https://laratranslate.com/translate), enabling professional translation capabilities with support for language detection, context-aware translations, translation memories, and glossaries.
  3. Lara leverages Translation Language Models (T-LMs) trained on billions of professionally translated segments, delivering domain-specific translations that capture cultural nuances and industry terminology that general-purpose LLMs often miss.
  4. [![License](https://img.shields.io/github/license/translated/lara-mcp.svg)](https://github.com/translated/lara-mcp/blob/main/LICENSE)
  5. [![Docker Pulls](https://img.shields.io/docker/pulls/translatednet/lara-mcp.svg)](https://hub.docker.com/r/translatednet/lara-mcp)
  6. [![npm downloads](https://img.shields.io/npm/dm/@translated/lara-mcp.svg)](https://www.npmjs.com/package/@translated/lara-mcp)
  7. ## Quick Start
  8. Pick your client below — no API keys needed, just log in through your browser.
  9. > The previous URL, https://mcp-v2.laratranslate.com/v1, keeps working: existing connectors need no change.
  10. ### Claude Desktop
  11. 1. Go to **Settings** > **Connectors**
  12. 2. Click **Add Custom Connector**
  13. 3. Enter the name: `Lara`
  14. 4. Enter the URL: `https://mcp.laratranslate.com/v1`
  15. 5. Click **Add**, then click **Connect**
  16. 6. Log in with your Lara Translate credentials in the browser
  17. Done — Lara Translate is now available in your conversations.
  18. ### Cursor
  19. Add to `~/.cursor/mcp.json`:
  20. ```json
  21. {
  22. "mcpServers": {
  23. "lara-translate": {
  24. "url": "https://mcp.laratranslate.com/v1"
  25. }
  26. }
  27. }
  28. ```
  29. Restart Cursor. The first time you use a Lara tool, your browser will open to authenticate.
  30. Full instructions: [Client Setup Guide](docs/client-setup.md#cursor).
  31. ### Claude Code
  32. ```bash
  33. claude mcp add --transport http lara-translate https://mcp.laratranslate.com/v1
  34. ```
  35. Then run `/mcp` inside Claude Code to log in through your browser.
  36. Full instructions: [Client Setup Guide](docs/client-setup.md#claude-code).
  37. ### Other Clients
  38. For step-by-step OAuth setup on **VS Code (GitHub Copilot)**, **Windsurf**, **Cline**, **Continue**, and more, see the **[Client Setup Guide](docs/client-setup.md)**.
  39. If your client isn't listed, the general approach is to add the server URL (`https://mcp.laratranslate.com/v1`) to your MCP config — the client will handle OAuth authentication automatically.
  40. > For a complete list of MCP-compatible clients, visit the [official MCP clients page](https://modelcontextprotocol.io/clients).
  41. ### Verify It Works
  42. After setup, test with a simple prompt:
  43. ```
  44. Translate with Lara "Hello world" to Spanish
  45. ```
  46. Your client should invoke Lara Translate and return the translation.
  47. ---
  48. ## Available Tools
  49. ### Translation
  50. | Tool | Description |
  51. |------|-------------|
  52. | `translate` | Translate text between languages with support for context, instructions, translation memories, glossaries, and multiple styles (faithful/fluid/creative) |
  53. ### Language Detection
  54. | Tool | Description |
  55. |------|-------------|
  56. | `detect_language` | Detect the language of a given text or array of texts |
  57. | `list_languages` | List all supported language codes |
  58. ### Translation Memories
  59. | Tool | Description |
  60. |------|-------------|
  61. | `list_memories` | List all translation memories in your account |
  62. | `create_memory` | Create a new translation memory |
  63. | `update_memory` | Update a translation memory's name |
  64. | `delete_memory` | Delete a translation memory |
  65. | `add_translation` | Add a translation unit (source + target pair) to a memory |
  66. | `delete_translation` | Delete a translation unit from a memory |
  67. | `import_tmx` | Import a TMX file into a memory |
  68. | `check_import_status` | Check the status of a TMX import job |
  69. ### Glossaries
  70. | Tool | Description |
  71. |------|-------------|
  72. | `list_glossaries` | List all glossaries in your account |
  73. | `get_glossary` | Get details of a specific glossary |
  74. | `create_glossary` | Create a new glossary |
  75. | `update_glossary` | Update a glossary's name |
  76. | `delete_glossary` | Delete a glossary |
  77. | `add_glossary_entry` | Add or replace a term entry in a glossary |
  78. | `delete_glossary_entry` | Delete a term entry from a glossary |
  79. | `import_glossary_csv` | Import entries from a CSV file into a glossary |
  80. | `check_glossary_import_status` | Check the status of a glossary CSV import job |
  81. | `export_glossary` | Export a glossary as CSV |
  82. | `get_glossary_counts` | Get the number of entries in a glossary |
  83. ---
  84. ## Authentication
  85. ### OAuth 2.0 (default)
  86. This is the method used in the [Quick Start](#quick-start) above. You provide only the server URL in your client config — no API keys needed. Your client handles the OAuth flow automatically: it opens your browser, you log in with your Lara Translate credentials, and you're connected.
  87. For per-client OAuth setup instructions, see the **[Client Setup Guide](docs/client-setup.md)**.
  88. ### Access Key (alternative)
  89. If you prefer to authenticate with API keys instead of browser login, you can pass your credentials directly in the client config. Get your **Access Key ID** and **Secret** from [Lara Translate](https://developers.laratranslate.com/docs/getting-started#step-3---configure-your-credentials).
  90. See the [Access Key section in the Client Setup Guide](docs/client-setup.md#alternative-access-key-authentication) for config examples.
  91. ---
  92. ## Self-Hosting
  93. Most users can connect to the hosted endpoint (`https://mcp.laratranslate.com/v1`) using the [Quick Start](#quick-start) instructions above. The options below are for running the server yourself.
  94. ### STDIO via NPX
  95. Requires [Node.js](https://nodejs.org/) 22.23.4 or later (the `engines` field of the published package). Only needed to run the server locally: the hosted endpoint above needs no installation.
  96. ```json
  97. {
  98. "mcpServers": {
  99. "lara-translate": {
  100. "command": "npx",
  101. "args": ["-y", "@translated/lara-mcp@latest"],
  102. "env": {
  103. "LARA_ACCESS_KEY_ID": "<YOUR_ACCESS_KEY_ID>",
  104. "LARA_ACCESS_KEY_SECRET": "<YOUR_ACCESS_KEY_SECRET>"
  105. }
  106. }
  107. }
  108. }
  109. ```
  110. ### STDIO via Docker
  111. Requires [Docker](https://www.docker.com/).
  112. ```json
  113. {
  114. "mcpServers": {
  115. "lara-translate": {
  116. "command": "docker",
  117. "args": [
  118. "run", "-i", "--rm",
  119. "-e", "LARA_ACCESS_KEY_ID",
  120. "-e", "LARA_ACCESS_KEY_SECRET",
  121. "translatednet/lara-mcp:latest"
  122. ],
  123. "env": {
  124. "LARA_ACCESS_KEY_ID": "<YOUR_ACCESS_KEY_ID>",
  125. "LARA_ACCESS_KEY_SECRET": "<YOUR_ACCESS_KEY_SECRET>"
  126. }
  127. }
  128. }
  129. }
  130. ```
  131. ### Building from Source
  132. #### Node.js
  133. Requires Node.js 22.23.4 or later and pnpm.
  134. ```bash
  135. git clone https://github.com/translated/lara-mcp.git
  136. cd lara-mcp
  137. pnpm install
  138. pnpm run build
  139. ```
  140. Then add to your MCP config:
  141. ```json
  142. {
  143. "mcpServers": {
  144. "lara-translate": {
  145. "command": "node",
  146. "args": ["<FULL_PATH_TO_PROJECT>/dist/index.js"],
  147. "env": {
  148. "LARA_ACCESS_KEY_ID": "<YOUR_ACCESS_KEY_ID>",
  149. "LARA_ACCESS_KEY_SECRET": "<YOUR_ACCESS_KEY_SECRET>"
  150. }
  151. }
  152. }
  153. }
  154. ```
  155. #### Docker
  156. ```bash
  157. git clone https://github.com/translated/lara-mcp.git
  158. cd lara-mcp
  159. docker build -t lara-mcp .
  160. ```
  161. Then add to your MCP config:
  162. ```json
  163. {
  164. "mcpServers": {
  165. "lara-translate": {
  166. "command": "docker",
  167. "args": [
  168. "run", "-i", "--rm",
  169. "-e", "LARA_ACCESS_KEY_ID",
  170. "-e", "LARA_ACCESS_KEY_SECRET",
  171. "lara-mcp"
  172. ],
  173. "env": {
  174. "LARA_ACCESS_KEY_ID": "<YOUR_ACCESS_KEY_ID>",
  175. "LARA_ACCESS_KEY_SECRET": "<YOUR_ACCESS_KEY_SECRET>"
  176. }
  177. }
  178. }
  179. }
  180. ```
  181. ---
  182. ## Usage Metrics
  183. The server reports usage events to Lara so we can see which tools and options
  184. are used and which problems users run into. Usage metrics are **on by default**.
  185. They are **pseudonymous**, not anonymous: each event is tied to your Lara account
  186. id.
  187. **Events:** an `auth_success` the first time your credentials are accepted, a
  188. `call_success` / `call_error` per tool call (`quota_exceeded` instead of
  189. `call_error` when the account is out of quota), and an `auth_fail` when Lara
  190. rejects your credentials (sent without an account id when the key never worked).
  191. **What is sent:**
  192. - your Lara account id (`acc_...`) and the server version;
  193. - the tool name, how long the call took, and how many characters were translated;
  194. - for `translate`: the source and target language tags, the options you used,
  195. and how many text blocks, glossaries, memories and instructions the call had
  196. (how many — never which);
  197. - the kind of client (`claude`, `claude-code`, `chatgpt`, `cursor`, `codex`,
  198. `vscode` or `other`) and its version number;
  199. - the transport (`stdio` / `http`), operating system and Node.js version;
  200. - on errors: an error category, the HTTP status and error type Lara answered
  201. with, and the names of invalid arguments.
  202. **What is never sent:** the text you translate or its translation, your access
  203. key id and secret, glossary and memory ids, instructions, context, error
  204. messages, raw client names or user agents.
  205. An installation id (a random UUID, tied to nothing) is generated the first time
  206. an event is sent and kept at `~/.lara/installation-id` — set `LARA_HOME` to move
  207. it. It identifies this installation to the metrics backend, never you.
  208. ### Opting out
  209. **Per call:** a `translate` call with `no_trace: true` (Lara's own privacy flag)
  210. reports nothing at all — no event of any kind is sent for that call.
  211. **Entirely:** set `DO_NOT_TRACK` to `1`, `true` or `yes` and nothing is
  212. measured, queued, written to disk or sent:
  213. ```json
  214. {
  215. "mcpServers": {
  216. "lara-translate": {
  217. "command": "npx",
  218. "args": ["-y", "@translated/lara-mcp@latest"],
  219. "env": {
  220. "LARA_ACCESS_KEY_ID": "<YOUR_ACCESS_KEY_ID>",
  221. "LARA_ACCESS_KEY_SECRET": "<YOUR_ACCESS_KEY_SECRET>",
  222. "DO_NOT_TRACK": "1"
  223. }
  224. }
  225. }
  226. }
  227. ```
  228. Metrics never block, slow down or fail a translation: events are batched, sent
  229. in the background, and a monitoring backend that is down is ignored.
  230. ---
  231. ## Support
  232. - For issues with Lara Translate API: visit [Lara Translate Support](https://support.laratranslate.com)
  233. - For issues with this MCP server: open an issue on [GitHub](https://github.com/translated/lara-mcp/issues)
  234. ## Lara Think
  235. Select Lara Think with `reasoning: true` on the `translate` tool. Choose Low,
  236. Medium or High with `reasoning_effort: "low" | "medium" | "high"`. Think defaults
  237. to **high** when the level is omitted. Without reasoning, Lara Base is used.
  238. A level requires `reasoning: true`; other combinations are rejected before the API call.
  239. ```json
  240. {"text":[{"text":"Hello","translatable":true}],"target":"it","reasoning":true,"reasoning_effort":"medium"}
  241. ```
  242. Think can take substantially longer. Existing timeout options still apply.
  243. When the MCP client supplies a `progressToken`, Lara Think streaming updates are
  244. forwarded as `notifications/progress`, without translation text or a percentage.
  245. These updates do not guarantee a longer client timeout; configure the client's
  246. request timeout separately.
  247. If Lara reports a reasoning fallback, the translation is retained and
  248. `warnings.reasoning.applied: false` and an explanatory text warning are returned.
  249. Absence of feedback does not prove which reasoning level ran internally.
  250. Usage metrics are enabled by default and can be disabled as described in Usage Metrics above.
  251. When enabled, they include the requested effective level and reported fallback, never source text.

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.