DeepLSAFE
A Model Context Protocol (MCP) server that provides translation capabilities using the DeepL API.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://www.npmjs.org/package/deepl-mcp-server) [](https://github.com/DeepL/deepl-mcp-server/blob/main/LICENSE)
A Model Context Protocol (MCP) server that exposes DeepL text translation, document translation, rephrasing, and glossary and style rule lookups as MCP tools.
Usage
You need Node.js 18 or newer and a DeepL API key. Pick an API plan and create a key in your DeepL account.
The server communicates over stdio, so it is normally started by an MCP client rather than by hand. To try it directly:
DEEPL_API_KEY=your-api-key npx -y deepl-mcp-server
It exits immediately if DEEPL_API_KEY is not set.
To work on the server itself:
git clone https://github.com/DeepL/deepl-mcp-server.git cd deepl-mcp-server npm install
Configuration
Register it as a stdio server with DEEPL_API_KEY in the server's environment. In Claude Code:
claude mcp add deepl --env DEEPL_API_KEY=your-api-key -- npx -y deepl-mcp-server
For clients configured through a JSON file:
{
"mcpServers": {
"deepl": {
"command": "npx",
"args": ["-y", "deepl-mcp-server"],
"env": {
"DEEPL_API_KEY": "your-api-key"
}
}
}
}Where that file lives and how each client expects local stdio servers to be declared:
To run a local checkout instead of the published package, use node as the command and the absolute path to src/index.mjs as the argument.
Tools
86122967faebOBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add deepl-mcp-server --env DEEPL_API_KEY=${DEEPL_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"deepl-mcp-server": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"DEEPL_API_KEY": "${DEEPL_API_KEY}"
}
}
}
}Exposed tools (13)
13 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
get-custom-instruction | read | Get a single custom instruction belonging to a style rule. Use the get-style-rule tool to find out which custom instructions a style rule contains. |
get-glossary-dictionary-entries | read | |
get-glossary-info | read | Given an id, get metadata about the glossary with that id - its name, available dictionaries, and creation time. |
get-source-languages | read | Get list of available source languages for translation |
get-style-rule | read | Given an id, get a single style rule with its full detail - configured rules and custom instructions. |
get-target-languages | read | Get list of available target languages for translation |
get-writing-styles | read | Get list of writing styles the DeepL API can use while rephrasing text |
get-writing-tones | read | Get list of writing tones the DeepL API can use while rephrasing text |
list-glossaries | read | Get a list of all glossaries with metadata for each - name, dictionaries available, and creation time. |
list-style-rules | read | Get a list of all style rules with metadata for each - id, name, language, and timestamps. Style rules can be applied when translating text or documents. Use the get-style-rule tool to fetch the configured rules and custom instructions of a single style rule. |
rephrase-text | read | Rephrase text in the same language, or into a different language, using DeepL API |
translate-document | read | Translate a document file using DeepL API |
translate-text | read |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (5)
.auto-approve.yaml
.gitlab-ci.yml
.prettierignore
.prettierrc.json
.releaserc.mjs
Gates applied: no_behavioural_pass.
86122967faebfull audit observations/trust-audit/mcp-server/deeplcom__deepl.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 86122967faeb | SAFE | B | 89 | first audit |
Questions
What is the DeepL MCP server?
A Model Context Protocol (MCP) server that provides translation capabilities using the DeepL API.
What tools does DeepL expose?
13 in total: 13 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is DeepL safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does DeepL need?
It reads DEEPL_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does DeepL run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as deepl-mcp-server at 1.3.12.
How current is this page?
The grade is for one exact copy of the source (86122967faeb), read on 2026-10-07. The repository is watched and re-audited when it changes.