Connectors · Marketing & media

DeepL MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to DeepL and the rest of your systems. It gives your agent 63 DeepL actions (28 reads, 25 writes and 10 that wait for a confirmation) and holds the DeepL credential itself, so the agent never does. It paces DeepL's limits and records every call.

Connect DeepL free See pricing

100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.

claude mcp add --transport http actionlake https://connect.actionlake.com/mcp

63actionsin one signed file
28readlook, never change
25writecreate and update
10wait for a yesdestructive, confirmed first

Set up

Connect DeepL in three steps.

01
Add actionlake to Claude.
Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
https://connect.actionlake.com/mcp
02
Connect DeepL.
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
- auth_keyDeepL authentication key (Account > API Keys)
- planYour DeepL plan
```
account { op: "connections.requirements", service: "deepl" }
# free: exactly what to paste, and what is found for you
```

Make the first call. Check before a write.

Ask in plain words; Claude picks the action. Before anything changes, check validates the call and prices it, for free.

do    { action: "deepl.admin_developer_keys.list" }# get all developer keys as an admin · read · 1 upstream callcheck { action: "deepl.text.translate", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

DeepL at a glance

Sign-in: API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
You provide: auth_key
Setup fields: plan
Limits we pace to: 50 calls a second, per connection
API covered: DeepL API OpenAPI 3.14.0, at your region's deepl.com address, set when you connect
Built from: DeepL's documentation, read 22 October 2026OpenAPI spec (GitHub)developers.deepl.com
Service file: Version 1.0.0, signed (Ed25519)
Cached reads: 24 of 28 reads, for up to 5 minutes; a write clears what it affects
Shared entities: 2, so ask can line DeepL up with your other systems
Data: Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in DeepL.

63 actions in 1 area. Every one carries its effect, so a key can allow the reads and hold the rest.

28readlook, never change
25writecreate and update
10destructivewait for a confirmation

Translation
63
28 read · 25 write · 10 destructive
Retrieve Languages · Get all developer keys as an admin · List all Glossaries · Request Translation

Top actions

The DeepL API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 2 of DeepL's 2 action names are shared with at least one other connector, so an agent learns them once.

Action	What it does	Effect	Same name in
`deepl.languages.list`Retrieve Languages	Retrieve Languages	read	`reference.language.list`8 connectors
`deepl.glossaries.list`List all Glossaries	List all Glossaries	read	this connector
`deepl.admin_developer_keys.list`Get all developer keys as an admin	Get all developer keys as an admin	read	this connector
`deepl.text.translate`Request Translation	Request Translation	write	`platform.translation.create`7 connectors
`deepl.style_rules.list`Retrieve style rule lists	Retrieve style rule lists	read	this connector
`deepl.glossaries.create`Create a Glossary	Create a Glossary	write	this connector
`deepl.languages_v2.list`Retrieve Supported Languages	Retrieve Supported Languages	read	this connector
`deepl.spoken_terms.list`List all Spoken Terms collections	List all Spoken Terms collections	read	this connector
`deepl.glossaries_v2.list`List all Glossaries	List all Glossaries	read	this connector
`deepl.style_rules.create`Create a style rule list	Create a style rule list	write	this connector
`deepl.spoken_terms.create`Create Spoken Terms collection	Create Spoken Terms collection	write	this connector
`deepl.glossaries.delete`Delete a Glossary	Delete a Glossary	destructive	this connector

Try it

Prompts to paste into Claude.

Each one names the DeepL actions it will call, so you know what it reads and what it changes before you send it.

Readread
Get all developer keys as an admin in DeepL and tell me what stands out.
callsdeepl.admin_developer_keys.list
Writewrite
Request Translation in DeepL. Show me exactly what you will send before you save it.
callsdeepl.text.translate

Governance for DeepL

Destructive actions wait for a yes.

10 DeepL actions are destructive. These are all of them:

Delete a Glossarydeepl.glossaries.delete
Delete a style rule listdeepl.style_rules.delete
Delete Spoken Terms collectiondeepl.spoken_terms.delete
Delete a Glossarydeepl.glossaries_v2.delete
Delete a style profiledeepl.style_profiles.delete
Delete a term listdeepl.spoken_term_lists.delete
Delete a translation memorydeepl.translation_memories.delete
Deletes the dictionary associated with the given language pair with the given glossary ID.deepl.glossary_dictionaries.delete
Delete a custom instructiondeepl.style_rule_custom_instructions.delete
Deactivate a developer key as an admindeepl.admin_developer_keys.deactivate

A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.

dodeepl.glossaries.deleteconfirm_policy: human

Delete a Glossary?

intent: the user asked me to delete a Glossary

run suspended · waiting for a named person · key deepl-agent cannot confirm itself

In the console, or from an emailed signed link.

a key for this connector

account {  op: "keys.create", name: "deepl-agent",  allowed_actions: ["deepl.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "deepl.glossaries.delete", approver: "[email protected]" },    { match: "deepl.style_rules.delete", approver: "[email protected]" },    { match: "deepl.spoken_terms.delete", approver: "[email protected]" }  ]}

Illustrative. Scoped to DeepL only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What DeepL costs on actionlake.

Every DeepL action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call DeepL refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

1upstream call per actionall 63 DeepL actions
12actions pageeach page of a long list is one call
24reads can be cacheda cache hit costs nothing
100,000calls a month, freepaid plans from US$49 a month for 250,000

See pricing

Action index

Every DeepL action.

63 actions by area: the id your agent calls, what it does, and its effect.

Translation63

deepl.admin_analytics.getGet usage statistics as an adminread
deepl.admin_analytics.get_custom_tagsGet custom tag usage statistics as an adminread
deepl.admin_developer_keys.createCreate a developer key as an adminwrite
deepl.admin_developer_keys.deactivateDeactivate a developer key as an admindestructive
deepl.admin_developer_keys.listGet all developer keys as an adminread
deepl.admin_developer_keys.renameRename a developer key as an adminwrite
deepl.admin_developer_keys.set_usage_limitsSet developer key usage limits as an adminwrite
deepl.documents.get_statusCheck Document Statusread
deepl.glossaries.createCreate a Glossarywrite
deepl.glossaries.deleteDelete a Glossarydestructive
deepl.glossaries.getRetrieve Glossary Detailsread
deepl.glossaries.get_entriesRetrieve Glossary Entriesread
deepl.glossaries.listList all Glossariesread
deepl.glossaries.updateEdit glossary detailswrite
deepl.glossaries_v2.createCreate a Glossarywrite
deepl.glossaries_v2.deleteDelete a Glossarydestructive
deepl.glossaries_v2.getRetrieve Glossary Detailsread
deepl.glossaries_v2.listList all Glossariesread
deepl.glossary_dictionaries.deleteDeletes the dictionary associated with the given language pair with the given glossary ID.destructive
deepl.glossary_dictionaries.replaceReplaces or creates a dictionary in the glossary with the specified entries.write
deepl.glossary_language_pairs.listList Language Pairs Supported by Glossariesread
deepl.language_resources.listRetrieve Language Resourcesread
deepl.languages.listRetrieve Languagesread
deepl.languages_v2.listRetrieve Supported Languagesread
deepl.quality_evaluations.getPoll for the evaluation resultread
deepl.spoken_term_lists.deleteDelete a term listdestructive
deepl.spoken_term_lists.replaceReplace or create a term listwrite
deepl.spoken_terms.createCreate Spoken Terms collectionwrite
deepl.spoken_terms.deleteDelete Spoken Terms collectiondestructive
deepl.spoken_terms.getRetrieve Spoken Terms collection detailsread
deepl.spoken_terms.get_entriesRetrieve Spoken Terms entriesread
deepl.spoken_terms.listList all Spoken Terms collectionsread
deepl.spoken_terms.updateEdit Spoken Terms collection detailswrite
deepl.style_profiles.createCreate a style profilewrite
deepl.style_profiles.deleteDelete a style profiledestructive
deepl.style_profiles.getRetrieve a style profileread
deepl.style_profiles.listList style profilesread
deepl.style_profiles.updateUpdate a style profilewrite
deepl.style_rule_custom_instructions.createCreate a custom instructionwrite
deepl.style_rule_custom_instructions.deleteDelete a custom instructiondestructive
deepl.style_rule_custom_instructions.getGet a custom instructionread
deepl.style_rule_custom_instructions.updateReplace a custom instructionwrite
deepl.style_rules.createCreate a style rule listwrite
deepl.style_rules.deleteDelete a style rule listdestructive
deepl.style_rules.getGet a style rule listread
deepl.style_rules.listRetrieve style rule listsread
deepl.style_rules.updateUpdate a style rule list's namewrite
deepl.style_rules.update_configured_rulesReplace configured rules for a style rule listwrite
deepl.text.translateRequest Translationwrite
deepl.translation_memories.deleteDelete a translation memorydestructive
deepl.translation_memories.exportExport a translation memorywrite
deepl.translation_memories.getRetrieve a translation memoryread
deepl.translation_memories.importImport a translation memorywrite
deepl.translation_memories.listList translation memoriesread
deepl.translation_memories.list_segmentsList translation memory segmentsread
deepl.translation_memory_jobs.getRetrieve an import or export jobread
deepl.usage.getCheck Usage and Limitsread
deepl.voice_sessions.createGet Streaming URLwrite
deepl.voice_sessions.reconnectRequest Reconnectionwrite
deepl.voice_translate_jobs.createCreate a voice translation jobwrite
deepl.voice_translate_jobs.getGet voice translation job statusread
deepl.write.correctCorrect textwrite
deepl.write.rephraseImprove textwrite

Questions

The DeepL MCP server, answered.

Is there a DeepL MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 63 DeepL actions, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by DeepL.

What do I need to connect DeepL?

DeepL authentication key (Account > API Keys). Plus your plan. The free call account { op: "connections.requirements", service: "deepl" } lists exactly what to paste.

Can an AI agent delete or change things in DeepL?

Only if its key allows it, and only with an explicit confirmation. 10 DeepL actions are destructive, including delete a Glossary, delete a style rule list and delete Spoken Terms collection. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.

Will an AI agent use up my DeepL API limit?

actionlake paces every call to the limits in the DeepL service file: 50 calls a second per connection. Repeated reads are answered from cache (24 DeepL actions can be), which never reaches DeepL.

What does a DeepL action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (12 DeepL actions page). Cache hits, searches, dry runs and calls DeepL refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.

Which DeepL API does it cover?

The DeepL API OpenAPI 3.14.0 at your region's deepl.com address, set when you connect. It is built from DeepL's documentation as read on 22 October 2026. 63 actions in 1 area, each one in the signed service file (version 1.0.0).

Does actionlake keep my DeepL data, and where?

actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your DeepL credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.

Which AI clients can use it?

Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.

Give your agent DeepL today.