Skip to content
actionlake
Start free

Connectors · AI

Groq MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Groq and the rest of your systems. It gives your agent 18 Groq actions (8 reads, 6 writes and 4 that wait for a confirmation) and holds the Groq credential itself, so the agent never does. It paces Groq's limits and records every call.

Connect Groq freeSee 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
the first call
do {  action: "groq.files.list",  intent: "check which Groq account is connected"}# Returns a list of files. · read · 1 upstream call
  • 18actionsin one signed file
  • 8readlook, never change
  • 6writecreate and update
  • 4wait for a yesdestructive, confirmed first

Set up

Connect Groq in three steps.

  1. 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
  2. 02

    Connect Groq.

    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.

    • api_keyGroq API key (console.groq.com > API Keys)
    account { op: "connections.requirements", service: "groq" }
# free: exactly what to paste, and what is found for you
  3. 03

    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: "groq.files.list" }# returns a list of files. · read · 1 upstream callcheck { action: "groq.batches.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Groq 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
api_key
Limits we pace to
60 calls a minute, per connection
API covered
GroqCloud API 2.1api.groq.com
Built from
Groq's documentation, read 23 October 2026storage.googleapis.comconsole.groq.com
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
8 of 8 reads, for up to 5 minutes; a write clears what it affects
Shared entities
1, so ask can line Groq up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Groq.

18 actions in 8 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 8readlook, never change
  • 6writecreate and update
  • 4destructivewait for a confirmation

  1. Batches

    4

    2 read · 1 write · 1 destructive

    List your organization's batches. · Creates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch). · Retrieves a batch. · Cancels a batch.

  2. Fine tunings

    4

    2 read · 1 write · 1 destructive

    Creates a new fine tuning for the already uploaded files This endpoint is in closed beta. [Contact us](https://groq.com/ · Lists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact) for m · Retrieves an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for more · Deletes an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for more i

  3. Files

    3

    2 read · 0 write · 1 destructive

    Returns a list of files. · Returns information about a file. · Delete a file.

  4. Models

    3

    2 read · 0 write · 1 destructive

    Get detailed information about a [model](https://console.groq.com/docs/models). · Delete model · List all available [models](https://console.groq.com/docs/models).

  5. Chat completions

    1

    0 read · 1 write · 0 destructive

    Creates a model response for the given chat conversation.

  6. Embeddings

    1

    0 read · 1 write · 0 destructive

    Creates an embedding vector representing the input text.

  7. Reranking

    1

    0 read · 1 write · 0 destructive

    Reranks documents based on their relevance to a query.

  8. Responses

    1

    0 read · 1 write · 0 destructive

    Creates a model response for the given input.

Top actions

The Groq API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 10 of Groq's 10 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
groq.files.listReturns a list of files.Returns a list of files.readcontent.file.list77 connectors
groq.files.getReturns information about a file.Returns information about a file.readcontent.file.get74 connectors
groq.batches.listList your organization's batches.List your organization's batches.readautomation.background_job.list34 connectors
groq.batches.createCreates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch).Creates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch).writeautomation.background_job.create29 connectors
groq.chat_completions.createCreates a model response for the given chat conversation.Creates a model response for the given chat conversation.writeanalytics.inference.create17 connectors
groq.models.getGet detailed information about a [model](https://console.groq.com/docs/models).Get detailed information about a [model](https://console.groq.com/docs/models).readinfra.ml_model.get13 connectors
groq.models.deleteDelete modelDelete modeldestructiveinfra.ml_model.delete9 connectors
groq.models.listList all available [models](https://console.groq.com/docs/models).List all available [models](https://console.groq.com/docs/models).readthis connector
groq.reranking.createReranks documents based on their relevance to a query.Reranks documents based on their relevance to a query.writethis connector
groq.responses.createCreates a model response for the given input.Creates a model response for the given input.writethis connector
groq.embeddings.createCreates an embedding vector representing the input text.Creates an embedding vector representing the input text.writethis connector
groq.fine_tunings.listLists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact) for mLists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact) for more information.readthis connector

Try it

Prompts to paste into Claude.

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

  1. Readread

    Which Groq files changed this week, and who changed them?

    callsgroq.files.list

  2. Writewrite

    Creates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch) in Groq. Show me exactly what you will send before you save it.

    callsgroq.batches.create

Governance for Groq

Destructive actions wait for a yes.

4 Groq actions are destructive. These are all of them:

  • Delete a file.groq.files.delete
  • Cancels a batch.groq.batches.cancel
  • Delete modelgroq.models.delete
  • Deletes an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for more igroq.fine_tunings.delete

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.

dogroq.files.deleteconfirm_policy: human

Delete a file.?

intent: the user asked me to delete a file.

run suspended · waiting for a named person · key groq-agent cannot confirm itself
In the console, or from an emailed signed link.
a key for this connector
account {  op: "keys.create", name: "groq-agent",  allowed_actions: ["groq.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "groq.files.delete", approver: "[email protected]" },    { match: "groq.batches.cancel", approver: "[email protected]" },    { match: "groq.models.delete", approver: "[email protected]" }  ]}
Illustrative. Scoped to Groq only, capped at 2,000 upstream calls, every write confirmed, and the destructive ones approved by a named person.

Cost

What Groq costs on actionlake.

Every Groq action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Groq 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 18 Groq actions
  • 4actions pageeach page of a long list is one call
  • 8reads 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 Groq action.

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

Batches4
  • groq.batches.cancelCancels a batch.destructive
  • groq.batches.createCreates and executes a batch from an uploaded file of requests. [Learn more](/docs/batch).write
  • groq.batches.getRetrieves a batch.read
  • groq.batches.listList your organization's batches.read
Fine tunings4
  • groq.fine_tunings.createCreates a new fine tuning for the already uploaded files This endpoint is in closed beta. [Contact us](https://groq.com/write
  • groq.fine_tunings.deleteDeletes an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for more idestructive
  • groq.fine_tunings.getRetrieves an existing fine tuning by id This endpoint is in closed beta. [Contact us](https://groq.com/contact) for moreread
  • groq.fine_tunings.listLists all previously created fine tunings. This endpoint is in closed beta. [Contact us](https://groq.com/contact) for mread
Files3
  • groq.files.deleteDelete a file.destructive
  • groq.files.getReturns information about a file.read
  • groq.files.listReturns a list of files.read
Models3
  • groq.models.deleteDelete modeldestructive
  • groq.models.getGet detailed information about a [model](https://console.groq.com/docs/models).read
  • groq.models.listList all available [models](https://console.groq.com/docs/models).read
Chat completions1
  • groq.chat_completions.createCreates a model response for the given chat conversation.write
Embeddings1
  • groq.embeddings.createCreates an embedding vector representing the input text.write
Reranking1
  • groq.reranking.createReranks documents based on their relevance to a query.write
Responses1
  • groq.responses.createCreates a model response for the given input.write

Questions

The Groq MCP server, answered.

Is there a Groq MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 18 Groq actions across batches, fine tunings, files and models, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Groq.

What do I need to connect Groq?

Groq API key (console.groq.com > API Keys). The free call account { op: "connections.requirements", service: "groq" } lists exactly what to paste.

Can an AI agent delete or change things in Groq?

Only if its key allows it, and only with an explicit confirmation. 4 Groq actions are destructive, including delete a file., cancels a batch. and delete model. 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 Groq API limit?

actionlake paces every call to the limits in the Groq service file: 60 calls a minute per connection. Repeated reads are answered from cache (8 Groq actions can be), which never reaches Groq.

What does a Groq action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (4 Groq actions page). Cache hits, searches, dry runs and calls Groq 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 Groq API does it cover?

The GroqCloud API 2.1 at api.groq.com. It is built from Groq's documentation as read on 23 October 2026. 18 actions in 8 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Groq 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 Groq 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.

More connectors like Groq

Shares the most action names with

Joins with, on shared entities

Browse

actionlake is independent. It is not made or endorsed by Groq; the name is the vendor's own. Counts are from the Groq service file, version 1.0.0.

Give your agent Groq today.

Connect Groq freeSee pricing

Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.