Connectors · Observability & analytics

Databox MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Databox and the rest of your systems. It gives your agent 88 Databox actions (52 reads, 27 writes and 9 that wait for a confirmation) and holds the Databox credential itself, so the agent never does. It paces Databox's limits and records every call.

Connect Databox 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

88actionsin one signed file
52readlook, never change
27writecreate and update
9wait for a yesdestructive, confirmed first

Set up

Connect Databox 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 Databox.
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_keyDatabox API key: in Databox open Profile > API keys > Create API key (paid plans)
```
account { op: "connections.requirements", service: "databox" }
# 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: "databox.users.list" }# list users · read · 1 upstream callcheck { action: "databox.users.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Databox 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: 100 calls a minute, per connection
API covered: Databox API v2 (OpenAPI 3.0.0)api.databox.com
Built from: Databox's documentation, read 23 October 2026developers.databox.com developers.databox.com
Service file: Version 1.0.0, signed (Ed25519)
Cached reads: 45 of 52 reads, for up to 5 minutes; a write clears what it affects
Shared entities: 4, so ask can line Databox up with your other systems
Data: Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Databox.

88 actions in 16 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

52readlook, never change
27writecreate and update
9destructivewait for a confirmation

Datasets
20
9 read · 9 write · 2 destructive
List datasets · Create a dataset · Retrieve a dataset · Update a dataset
Data sources
11
4 read · 5 write · 2 destructive
Create a data source · List data sources · Retrieve a data source's available sync frequencies · Retrieve a data source
Metrics
11
7 read · 3 write · 1 destructive
List metrics · Create a metric · Update a metric · Retrieve a metric
Organization
7
6 read · 1 write · 0 destructive
List activity log entries · List countries · List metadata options · List time zones
Connections
6
3 read · 2 write · 1 destructive
List connections · Retrieve a connection · Retrieve a connection's permissions · Update a connection
Dataset modifications
6
4 read · 1 write · 1 destructive
List dataset modification functions · Retrieve dataset modification rules · Preview a dataset modification · Retrieve a dataset's modification
Accounts
5
2 read · 2 write · 1 destructive
Create an account · List accounts · Retrieve an account · Update an account
Users
5
2 read · 2 write · 1 destructive
List users · Create a user · Retrieve a user · Update a user

8 more areas · 17 actions

Dataset ingestions 3
Profile 3
Billing 2
Databoards 2
Dataset data 2
Dataset syncs 2
Integrations 2
Auth 1

Top actions

The Databox API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 17 of Databox's 17 action names are shared with at least one other connector, so an agent learns them once.

Action	What it does	Effect	Same name in
`databox.users.list`List users	List users.	read	`identity.user.list`474 connectors
`databox.profile.get`Retrieve profile	Retrieve profile.	read	`identity.current_user.get`307 connectors
`databox.datasets.list`List datasets	List datasets.	read	`analytics.dataset.list`28 connectors
`databox.users.create`Create a user	Create a user.	write	`identity.user.create`262 connectors
`databox.datasets.create`Create a dataset	Create a dataset.	write	`analytics.dataset.create`21 connectors
`databox.users.delete`Delete a user	Delete a user.	destructive	`identity.user.delete`253 connectors
`databox.databoards.list`List databoards	List databoards.	read	`analytics.dashboard.list`22 connectors
`databox.metrics.list`List metrics	List metrics.	read	`analytics.metric_definition.list`11 connectors
`databox.metrics.create`Create a metric	Create a metric.	write	`analytics.metric_definition.create`10 connectors
`databox.data_sources.list`List data sources	List data sources.	read	this connector
`databox.datasets.duplicate`Duplicate a dataset	Duplicate a dataset.	write	this connector
`databox.data_sources.create`Create a data source	Create a data source.	write	this connector

Try it

Prompts to paste into Claude.

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

Readread
Who has access in Databox? List every user and their role.
callsdatabox.users.list
Writewrite
Create a user in Databox. Show me exactly what you will send before you save it.
callsdatabox.users.create

Governance for Databox

Destructive actions wait for a yes.

9 Databox actions are destructive. These are all of them:

Delete a userdatabox.users.delete
Delete a datasetdatabox.datasets.delete
Delete a metricdatabox.metrics.delete
Purge a datasetdatabox.datasets.purge
Purge a data sourcedatabox.data_sources.purge
Delete a data sourcedatabox.data_sources.delete
Delete an accountdatabox.accounts.delete
Delete a connectiondatabox.connections.delete
Delete a dataset modificationdatabox.dataset_modifications.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.

dodatabox.users.deleteconfirm_policy: human

Delete a user?

intent: the user asked me to delete a user

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

In the console, or from an emailed signed link.

a key for this connector

account {  op: "keys.create", name: "databox-agent",  allowed_actions: ["databox.*"], cap_calls: 2000,  confirm_policy: "human_for_writes",  approval_rules: [    { match: "databox.users.delete", approver: "[email protected]" },    { match: "databox.datasets.delete", approver: "[email protected]" },    { match: "databox.metrics.delete", approver: "[email protected]" }  ]}

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

Cost

What Databox costs on actionlake.

Every Databox action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Databox 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 88 Databox actions
88actions pageeach page of a long list is one call
45reads 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 Databox action.

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

Datasets20

databox.datasets.createCreate a datasetwrite
databox.datasets.deleteDelete a datasetdestructive
databox.datasets.duplicateDuplicate a datasetwrite
databox.datasets.getRetrieve a datasetread
databox.datasets.get_column_metadataRetrieve a dataset's column metadataread
databox.datasets.get_lineageRetrieve a dataset's lineageread
databox.datasets.get_metadataRetrieve a dataset's metadataread
databox.datasets.get_permissionsRetrieve a dataset's permissionsread
databox.datasets.get_schemaRetrieve a dataset's schemaread
databox.datasets.get_verificationRetrieve a dataset's verificationread
databox.datasets.listList datasetsread
databox.datasets.list_sync_frequency_optionsRetrieve a dataset's sync frequency optionsread
databox.datasets.purgePurge a datasetdestructive
databox.datasets.updateUpdate a datasetwrite
databox.datasets.update_column_metadataUpdate a dataset's column metadatawrite
databox.datasets.update_metadataUpdate a dataset's metadatawrite
databox.datasets.update_permissionsUpdate a dataset's permissionswrite
databox.datasets.update_sync_frequencyUpdate a dataset's sync frequencywrite
databox.datasets.update_timezoneUpdate a dataset's time zonewrite
databox.datasets.update_verificationUpdate a dataset's verificationwrite

Data sources11

databox.data_sources.createCreate a data sourcewrite
databox.data_sources.deleteDelete a data sourcedestructive
databox.data_sources.getRetrieve a data sourceread
databox.data_sources.get_permissionsRetrieve a data source's permissionsread
databox.data_sources.listList data sourcesread
databox.data_sources.list_sync_frequency_optionsRetrieve a data source's available sync frequenciesread
databox.data_sources.purgePurge a data sourcedestructive
databox.data_sources.updateUpdate a data sourcewrite
databox.data_sources.update_permissionsUpdate a data source's permissionswrite
databox.data_sources.update_sync_frequencyUpdate a data source's sync frequencywrite
databox.data_sources.update_timezoneUpdate a data source's time zonewrite

Metrics11

databox.metrics.createCreate a metricwrite
databox.metrics.deleteDelete a metricdestructive
databox.metrics.getRetrieve a metricread
databox.metrics.get_lineageRetrieve a metric's lineageread
databox.metrics.get_verificationRetrieve a metric's verificationread
databox.metrics.listList metricsread
databox.metrics.list_usagesRetrieve a metric's usagesread
databox.metrics.query_dimension_valuesRetrieve dimension valuesread
databox.metrics.query_drilldownRetrieve drilldown dataread
databox.metrics.updateUpdate a metricwrite
databox.metrics.update_verificationUpdate a metric's verificationwrite

Organization7

databox.organization.getRetrieve organization detailsread
databox.organization.get_usageRetrieve organization usageread
databox.organization.list_activity_logList activity log entriesread
databox.organization.list_countriesList countriesread
databox.organization.list_metadata_optionsList metadata optionsread
databox.organization.list_timezonesList time zonesread
databox.organization.updateUpdate organization detailswrite

Connections6

databox.connections.deleteDelete a connectiondestructive
databox.connections.getRetrieve a connectionread
databox.connections.get_permissionsRetrieve a connection's permissionsread
databox.connections.listList connectionsread
databox.connections.updateUpdate a connectionwrite
databox.connections.update_permissionsUpdate a connection's permissionswrite

Dataset modifications6

databox.dataset_modifications.deleteDelete a dataset modificationdestructive
databox.dataset_modifications.getRetrieve a dataset's modificationread
databox.dataset_modifications.list_functionsList dataset modification functionsread
databox.dataset_modifications.list_rulesRetrieve dataset modification rulesread
databox.dataset_modifications.previewPreview a dataset modificationread
databox.dataset_modifications.updateUpdate a dataset modificationwrite

Accounts5

databox.accounts.createCreate an accountwrite
databox.accounts.deleteDelete an accountdestructive
databox.accounts.getRetrieve an accountread
databox.accounts.listList accountsread
databox.accounts.updateUpdate an accountwrite

Users5

databox.users.createCreate a userwrite
databox.users.deleteDelete a userdestructive
databox.users.getRetrieve a userread
databox.users.listList usersread
databox.users.updateUpdate a userwrite

Dataset ingestions3

databox.dataset_ingestions.getRetrieve a dataset ingestionread
databox.dataset_ingestions.get_statisticsRetrieve a dataset's ingestion statisticsread
databox.dataset_ingestions.listList a dataset's ingestionsread

Profile3

databox.profile.getRetrieve profileread
databox.profile.list_metadata_optionsList metadata optionsread
databox.profile.updateUpdate profilewrite

Billing2

databox.billing.getRetrieve billing detailsread
databox.billing.list_invoicesList invoicesread

Databoards2

databox.databoards.listList databoardsread
databox.databoards.list_metricsRetrieve a Databoard's metricsread

Dataset data2

databox.dataset_data.ingestIngest data into a datasetwrite
databox.dataset_data.listRetrieve dataset dataread

Dataset syncs2

databox.dataset_syncs.get_statisticsRetrieve a dataset's sync statisticsread
databox.dataset_syncs.listRetrieve a dataset's sync historyread

Integrations2

databox.integrations.getRetrieve an integrationread
databox.integrations.listList integrationsread

Auth1

databox.auth.validate_keyValidate an API keyread

Questions

The Databox MCP server, answered.

Is there a Databox MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 88 Databox actions across datasets, data sources, metrics and organization, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Databox.

What do I need to connect Databox?

Databox API key: in Databox open Profile > API keys > Create API key (paid plans). The free call account { op: "connections.requirements", service: "databox" } lists exactly what to paste.

Can an AI agent delete or change things in Databox?

Only if its key allows it, and only with an explicit confirmation. 9 Databox actions are destructive, including delete a user, delete a dataset and delete a metric. 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 Databox API limit?

actionlake paces every call to the limits in the Databox service file: 100 calls a minute per connection. Repeated reads are answered from cache (45 Databox actions can be), which never reaches Databox.

What does a Databox action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (88 Databox actions page). Cache hits, searches, dry runs and calls Databox 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 Databox API does it cover?

The Databox API v2 (OpenAPI 3.0.0) at api.databox.com. It is built from Databox's documentation as read on 23 October 2026. 88 actions in 16 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Databox 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 Databox 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 Databox today.

Connect Databox free See pricing

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

Databox MCP server for AI agents.

Connect Databox in three steps.

Add actionlake to Claude.

Connect Databox.

Make the first call. Check before a write.

What an AI agent can do in Databox.

Datasets

Data sources

Metrics

Organization

Connections

Dataset modifications

Accounts

Users