Skip to content
actionlake
Start free

Connectors · Identity & security

Torq MCP server for AI agents.

actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Torq and the rest of your systems. It gives your agent 118 Torq actions (56 reads, 46 writes and 16 that wait for a confirmation) and holds the Torq credential itself, so the agent never does. It paces Torq's limits and records every call.

Connect Torq 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: "torq.users.list",  intent: "check which Torq account is connected"}# List Users · read · 1 upstream call
  • 118actionsin one signed file
  • 56readlook, never change
  • 46writecreate and update
  • 16wait for a yesdestructive, confirmed first

Set up

Connect Torq 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 Torq.

    OAuth 2.0, client credentials. Paste the app's client id and secret; actionlake fetches and renews the token, and nobody has to sign in. actionlake stores what you paste write-only, sealed under your tenant's key.

    • client_idTorq API key Client ID
    • client_secretTorq API key Client Secret (shown once when the key is created)
    • tenant_hostWhere your Torq lives: your tenant host
    account { op: "connections.requirements", service: "torq" }
# 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: "torq.users.list" }# list users · read · 1 upstream callcheck { action: "torq.roles.create", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent

Torq at a glance

Sign-in
OAuth 2.0, client credentials. Paste the app's client id and secret; actionlake fetches and renews the token, and nobody has to sign in.
You provide
client_idclient_secret
Setup fields
tenant_host
Limits we pace to
50 calls a second, per connection
API covered
Torq public REST API (v1alpha + v1 triage)auth.torq.io/public
Built from
Torq's documentation, read 23 October 2026developers.torq.iodevelopers.torq.io
Service file
Version 1.0.0, signed (Ed25519)
Cached reads
51 of 56 reads, for up to 5 minutes; a write clears what it affects
Shared entities
2, so ask can line Torq up with your other systems
Data
Processed and stored in Sydney, Australia. No US region yet

Capabilities

What an AI agent can do in Torq.

118 actions in 34 areas. Every one carries its effect, so a key can allow the reads and hold the rest.

  • 56readlook, never change
  • 46writecreate and update
  • 16destructivewait for a confirmation

  1. Workflows

    13

    5 read · 6 write · 2 destructive

    List workflows · Retrieve a workflow · Update a workflow · Delete a workflow

  2. Cases

    10

    4 read · 4 write · 2 destructive

    Query cases · Create a case · Retrieve a case · Update a case

  3. Secrets

    6

    2 read · 3 write · 1 destructive

    List secrets · Create a secret · Retrieve a secret · Update a secret

  4. Triage alerts

    6

    3 read · 3 write · 0 destructive

    List alerts · Retrieve alert · List alert history · Change verdict

  5. Workspace variables

    6

    2 read · 3 write · 1 destructive

    Create a workspace variable · List workspace variables · Retrieve a workspace variable · Remove workspace variable sharing

  6. Case links

    5

    2 read · 2 write · 1 destructive

    Link cases · List the case links · List link relations · Update a link

  7. Case notes

    5

    2 read · 2 write · 1 destructive

    List the case notes · Add a note to a case · Retrieve a note · Update a note

  8. Roles

    5

    2 read · 2 write · 1 destructive

    List roles · Create a role · Retrieve a role · Update a role

26 more areas · 62 actions
  • SSO claim mappings 5
  • Step runners 5
  • Triage guidance 5
  • Case custom fields 4
  • Observables 4
  • Share requests 4
  • Users 4
  • Case observables 3
  • Case timeline 3
  • Runbooks 3
  • Workspaces 3
  • Executions 2
  • Integrations 2
  • Security settings 2
  • Triage verdict actions 2
  • Activity logs 1
  • API keys 1
  • Audit logs 1
  • Case assignees 1
  • Case attachments 1
  • Case categories 1
  • Case events 1
  • Case resolution reasons 1
  • Case reviewers 1
  • Case tags 1
  • Workflow tags 1

Top actions

The Torq API, shaped for agents.

The ids your agent types, what each does, and the one name it shares with other systems. 33 of Torq's 33 action names are shared with at least one other connector, so an agent learns them once.

ActionWhat it doesEffectSame name in
torq.users.listList UsersList Users.readidentity.user.list474 connectors
torq.roles.listList rolesList roles.readidentity.role.list184 connectors
torq.roles.createCreate a roleCreate a role.writeidentity.role.create105 connectors
torq.users.deleteDelete a userDelete a user.destructiveidentity.user.delete253 connectors
torq.roles.updateUpdate a roleUpdate a role.writeidentity.role.update98 connectors
torq.audit_logs.listList audit logsList audit logs.readsecurity.audit_log.list69 connectors
torq.api_keys.listList API keysList API keys.readidentity.api_key.list96 connectors
torq.triage_alerts.queryList alertsList alerts.readsecurity.alert.list22 connectors
torq.secrets.listList secretsList secrets.readsecurity.secret.list21 connectors
torq.secrets.createCreate a secretCreate a secret.writesecurity.secret.create19 connectors
torq.cases.queryQuery casesQuery cases.readsecurity.incident.list17 connectors
torq.case_notes.listList the case notesList the case notes.readsecurity.incident_note.list9 connectors

Try it

Prompts to paste into Claude.

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

  1. Readread

    Who has access in Torq? List every user and their role.

    callstorq.users.list

  2. Writewrite

    Create a role in Torq. Show me exactly what you will send before you save it.

    callstorq.roles.create

Governance for Torq

Destructive actions wait for a yes.

16 Torq actions are destructive. These are all of them:

  • Delete a usertorq.users.delete
  • Delete a roletorq.roles.delete
  • Delete a secrettorq.secrets.delete
  • Delete a notetorq.case_notes.delete
  • Delete a casetorq.cases.delete
  • Delete a workflowtorq.workflows.delete
  • Unlink casestorq.case_links.delete
  • Remove a tag from a casetorq.cases.remove_tags
  • Delete a step runnertorq.step_runners.delete
  • Remove a tagtorq.workflows.remove_tag
  • Deactivate a workspacetorq.workspaces.deactivate
  • Delete guidancetorq.triage_guidance.delete
  • Remove a custom field from a casetorq.case_custom_fields.remove
  • Delete a claim mappingtorq.sso_claim_mappings.delete
  • Delete a workspace variabletorq.workspace_variables.delete
  • Disassociate an observable from a casetorq.case_observables.disassociate

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.

dotorq.users.deleteconfirm_policy: human

Delete a user?

intent: the user asked me to delete a user

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

Cost

What Torq costs on actionlake.

Every Torq action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Torq 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 118 Torq actions
  • 118actions pageeach page of a long list is one call
  • 51reads 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 Torq action.

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

Workflows13
  • torq.workflows.add_tagAdd a tagwrite
  • torq.workflows.deleteDelete a workflowdestructive
  • torq.workflows.getRetrieve a workflowread
  • torq.workflows.get_revisionRetrieve a revisionread
  • torq.workflows.importImport a workflowwrite
  • torq.workflows.listList workflowsread
  • torq.workflows.list_parentsList parent workflowsread
  • torq.workflows.list_tagsList workflow tagsread
  • torq.workflows.publishPublish a workflowwrite
  • torq.workflows.remove_sharingRemove workflow sharingwrite
  • torq.workflows.remove_tagRemove a tagdestructive
  • torq.workflows.unpublishUnpublish a workflowwrite
  • torq.workflows.updateUpdate a workflowwrite
Cases10
  • torq.cases.add_tagsAdd a tag to a casewrite
  • torq.cases.bulk_createBulk create extended caseswrite
  • torq.cases.createCreate a casewrite
  • torq.cases.deleteDelete a casedestructive
  • torq.cases.getRetrieve a caseread
  • torq.cases.get_bulk_create_operationRetrieve bulk case creation stateread
  • torq.cases.list_bulk_create_operationsList bulk case creation stateread
  • torq.cases.queryQuery casesread
  • torq.cases.remove_tagsRemove a tag from a casedestructive
  • torq.cases.updateUpdate a casewrite
Secrets6
  • torq.secrets.createCreate a secretwrite
  • torq.secrets.deleteDelete a secretdestructive
  • torq.secrets.getRetrieve a secretread
  • torq.secrets.listList secretsread
  • torq.secrets.remove_sharingRemove secret sharingwrite
  • torq.secrets.updateUpdate a secretwrite
Triage alerts6
  • torq.triage_alerts.change_verdictChange verdictwrite
  • torq.triage_alerts.confirm_verdictConfirm verdictwrite
  • torq.triage_alerts.getRetrieve alertread
  • torq.triage_alerts.list_historyList alert historyread
  • torq.triage_alerts.queryList alertsread
  • torq.triage_alerts.undo_verdict_reviewUndo verdict reviewwrite
Workspace variables6
  • torq.workspace_variables.createCreate a workspace variablewrite
  • torq.workspace_variables.deleteDelete a workspace variabledestructive
  • torq.workspace_variables.getRetrieve a workspace variableread
  • torq.workspace_variables.listList workspace variablesread
  • torq.workspace_variables.remove_sharingRemove workspace variable sharingwrite
  • torq.workspace_variables.updateUpdate a workspace variablewrite
Case links5
  • torq.case_links.createLink caseswrite
  • torq.case_links.deleteUnlink casesdestructive
  • torq.case_links.listList the case linksread
  • torq.case_links.list_relationsList link relationsread
  • torq.case_links.updateUpdate a linkwrite
Case notes5
  • torq.case_notes.createAdd a note to a casewrite
  • torq.case_notes.deleteDelete a notedestructive
  • torq.case_notes.getRetrieve a noteread
  • torq.case_notes.listList the case notesread
  • torq.case_notes.updateUpdate a notewrite
Roles5
  • torq.roles.createCreate a rolewrite
  • torq.roles.deleteDelete a roledestructive
  • torq.roles.getRetrieve a roleread
  • torq.roles.listList rolesread
  • torq.roles.updateUpdate a rolewrite
SSO claim mappings5
  • torq.sso_claim_mappings.createAdd a claim mappingwrite
  • torq.sso_claim_mappings.deleteDelete a claim mappingdestructive
  • torq.sso_claim_mappings.getRetrieve a claim mappingread
  • torq.sso_claim_mappings.listList claim mappingsread
  • torq.sso_claim_mappings.updateUpdate a claim mappingwrite
Step runners5
  • torq.step_runners.createCreate a step runnerwrite
  • torq.step_runners.deleteDelete a step runnerdestructive
  • torq.step_runners.getRetrieve a step runnerread
  • torq.step_runners.listList step runnersread
  • torq.step_runners.regenerate_install_commandRegenerate a step runner install commandwrite
Triage guidance5
  • torq.triage_guidance.createCreate guidancewrite
  • torq.triage_guidance.deleteDelete guidancedestructive
  • torq.triage_guidance.getRetrieve guidanceread
  • torq.triage_guidance.listList guidanceread
  • torq.triage_guidance.updateUpdate guidancewrite
Case custom fields4
  • torq.case_custom_fields.addAdd a custom field to a casewrite
  • torq.case_custom_fields.listList custom fields for casesread
  • torq.case_custom_fields.removeRemove a custom field from a casedestructive
  • torq.case_custom_fields.setUpdate the value of a custom fieldwrite
Observables4
  • torq.observables.createCreate an observablewrite
  • torq.observables.getRetrieve an observableread
  • torq.observables.queryQuery observablesread
  • torq.observables.updateUpdate an observablewrite
Share requests4
  • torq.share_requests.acceptAccept a share requestwrite
  • torq.share_requests.createCreate a share requestwrite
  • torq.share_requests.denyDeny a share requestwrite
  • torq.share_requests.listList share requestsread
Users4
  • torq.users.deleteDelete a userdestructive
  • torq.users.inviteInvite a new userwrite
  • torq.users.listList Usersread
  • torq.users.list_rolesList rolesread
Case observables3
  • torq.case_observables.associateAssociate an observable with a casewrite
  • torq.case_observables.disassociateDisassociate an observable from a casedestructive
  • torq.case_observables.listList the case observablesread
Case timeline3
  • torq.case_timeline.add_commentAdd a timeline commentwrite
  • torq.case_timeline.attach_eventsAdd a timeline eventwrite
  • torq.case_timeline.queryQuery the case timelineread
Runbooks3
  • torq.runbooks.getRetrieve a runbook by IDread
  • torq.runbooks.get_by_nameRetrieve a runbook by nameread
  • torq.runbooks.listList runbooksread
Workspaces3
  • torq.workspaces.createCreate a workspacewrite
  • torq.workspaces.deactivateDeactivate a workspacedestructive
  • torq.workspaces.listList workspacesread
Executions2
  • torq.executions.getRetrieve an executionread
  • torq.executions.get_trigger_eventRetrieve a trigger eventread
Integrations2
  • torq.integrations.listList integrationsread
  • torq.integrations.remove_sharingRemove integration sharingwrite
Security settings2
  • torq.security_settings.getRetrieve security settingsread
  • torq.security_settings.setSet security settingswrite
Triage verdict actions2
  • torq.triage_verdict_actions.getRetrieve verdict actionsread
  • torq.triage_verdict_actions.updateUpdate verdict actionswrite
Activity logs1
  • torq.activity_logs.listList activity logsread
API keys1
  • torq.api_keys.listList API keysread
Audit logs1
  • torq.audit_logs.listList audit logsread
Case assignees1
  • torq.case_assignees.listList cases assigneesread
Case attachments1
  • torq.case_attachments.listList the case attachmentsread
Case categories1
  • torq.case_categories.listList cases categoriesread
Case events1
  • torq.case_events.queryList the case eventsread
Case resolution reasons1
  • torq.case_resolution_reasons.listList cases resolution reasonsread
Case reviewers1
  • torq.case_reviewers.listList case reviewersread
Case tags1
  • torq.case_tags.listList cases tagsread
Workflow tags1
  • torq.workflow_tags.listList tagsread

Questions

The Torq MCP server, answered.

Is there a Torq MCP server for Claude?

Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 118 Torq actions across workflows, cases, secrets and triage alerts, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Torq.

What do I need to connect Torq?

Torq API key Client ID and Torq API key Client Secret (shown once when the key is created). Plus your tenant host. The free call account { op: "connections.requirements", service: "torq" } lists exactly what to paste.

Can an AI agent delete or change things in Torq?

Only if its key allows it, and only with an explicit confirmation. 16 Torq actions are destructive, including delete a user, delete a role and delete a secret. 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 Torq API limit?

actionlake paces every call to the limits in the Torq service file: 50 calls a second per connection. Repeated reads are answered from cache (51 Torq actions can be), which never reaches Torq.

What does a Torq action cost?

One upstream call, the only unit actionlake bills. A list that pages bills one call per page (118 Torq actions page). Cache hits, searches, dry runs and calls Torq 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 Torq API does it cover?

The Torq public REST API (v1alpha + v1 triage) at auth.torq.io/public by default, or the base URL set when you connect. It is built from Torq's documentation as read on 23 October 2026. 118 actions in 34 areas, each one in the signed service file (version 1.0.0).

Does actionlake keep my Torq 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 Torq 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 Torq

Shares the most action names with

Joins with, on shared entities

Browse

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

Give your agent Torq today.

Connect Torq freeSee pricing

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