Connectors · MSP & PSA
Tailscale MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Tailscale and the rest of your systems. It gives your agent 93 Tailscale actions (38 reads, 40 writes and 15 that wait for a confirmation) and holds the Tailscale credential itself, so the agent never does. It paces Tailscale's limits and records every call.
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/mcpdo { action: "tailscale.users.list", intent: "check which Tailscale account is connected"}# List users · read · 1 upstream call- 93actionsin one signed file
- 38readlook, never change
- 40writecreate and update
- 15wait for a yesdestructive, confirmed first
Set up
Connect Tailscale 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 Tailscale.
Bearer token. 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.
access_tokenTailscale API access token (tskey-api-...): admin console > Settings > Keys > Generate access token
account { op: "connections.requirements", service: "tailscale" } # free: exactly what to paste, and what is found for you - 03
Make the first call. Check before a write.
Ask in plain words; Claude picks the action. Before anything changes,
checkvalidates the call and prices it, for free.do { action: "tailscale.users.list" }# list users · read · 1 upstream callcheck { action: "tailscale.users.delete", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Tailscale at a glance
- Sign-in
- Bearer token. 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
access_token- Limits we pace to
- 300 calls a minute, per connection
- API covered
- Tailscale API v2 (OpenAPI 3.1)
api.tailscale.com/api/v2 - Built from
- Tailscale's documentation, read 23 October 2026api.tailscale.comtailscale.com
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 35 of 38 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 2, so ask can line Tailscale up with your other systems
- Data
- Processed and stored in Sydney, Australia. No US region yet
Capabilities
What an AI agent can do in Tailscale.
93 actions in 26 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 38readlook, never change
- 40writecreate and update
- 15destructivewait for a confirmation
Devices
92 read · 5 write · 2 destructive
List tailnet devices · Get a device · Delete a device · Authorize device
Users
72 read · 3 write · 2 destructive
List users · Get a user · Delete a user · Approve a user
Webhooks
72 read · 3 write · 2 destructive
Create a webhook · List webhooks · Update webhook · Get webhook
Device invites
62 read · 3 write · 1 destructive
Create device invites · List device invites · Accept a device invite · Get a device invite
Keys
52 read · 2 write · 1 destructive
Create an auth key or trust credential · List tailnet keys · Get key · Delete key
Oauth apps
52 read · 2 write · 1 destructive
Create an OAuth app · List OAuth apps · Get OAuth app · Update OAuth app
Posture integrations
52 read · 2 write · 1 destructive
Create a posture integration · List all posture integrations · Get a posture integration · Delete a posture integration
User invites
52 read · 2 write · 1 destructive
List user invites · Create user invites · Delete a user invite · Get a user invite
18 more areas · 44 actions
- Device posture attributes 4
- Log streaming 4
- Policy file 4
- Services 4
- Contacts 3
- Split DNS 3
- Tailnets 3
- Aws external ids 2
- Device routes 2
- DNS configuration 2
- DNS nameservers 2
- DNS preferences 2
- DNS search paths 2
- Service device approvals 2
- Tailnet settings 2
- Configuration audit logs 1
- Network flow logs 1
- Service hosts 1
Top actions
The Tailscale API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 25 of Tailscale's 25 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
tailscale.users.listList users | List users. | read | identity.user.list474 connectors |
tailscale.users.getGet a user | Get a user. | read | identity.user.get399 connectors |
tailscale.users.deleteDelete a user | Delete a user. | destructive | identity.user.delete253 connectors |
tailscale.devices.listList tailnet devices | List tailnet devices. | read | device.endpoint.list65 connectors |
tailscale.keys.createCreate an auth key or trust credential | Create an auth key or trust credential. | write | identity.api_key.create100 connectors |
tailscale.keys.listList tailnet keys | List tailnet keys. | read | identity.api_key.list96 connectors |
tailscale.devices.getGet a device | Get a device. | read | device.endpoint.get57 connectors |
tailscale.webhooks.createCreate a webhook | Create a webhook. | write | automation.webhook.create209 connectors |
tailscale.webhooks.listList webhooks | List webhooks. | read | automation.webhook.list202 connectors |
tailscale.policy_file.getGet policy file | Get policy file. | read | security.policy.get30 connectors |
tailscale.policy_file.setSet policy file | Set policy file. | write | security.policy.update25 connectors |
tailscale.tailnets.createCreate a tailnet | Create a tailnet. | write | platform.tenant.create35 connectors |
Try it
Prompts to paste into Claude.
Each one names the Tailscale actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Which Tailscale devices have not checked in for a week?
calls
tailscale.devices.list - Writewrite
Create an auth key or trust credential in Tailscale. Show me exactly what you will send before you save it.
calls
tailscale.keys.create
Governance for Tailscale
Destructive actions wait for a yes.
15 Tailscale actions are destructive. These are all of them:
- Delete a user
tailscale.users.delete - Delete key
tailscale.keys.delete - Delete a device
tailscale.devices.delete - Delete webhook
tailscale.webhooks.delete - Delete a user invite
tailscale.user_invites.delete - Delete a tailnet
tailscale.tailnets.delete - Suspend a user
tailscale.users.suspend - Expire a device's key
tailscale.devices.expire_key - Delete a Service
tailscale.services.delete - Delete OAuth app
tailscale.oauth_apps.delete - Delete a device invite
tailscale.device_invites.delete - Disable log streaming
tailscale.log_streaming.disable - Rotate webhook secret
tailscale.webhooks.rotate_secret - Delete a posture integration
tailscale.posture_integrations.delete - Delete custom device posture attributes
tailscale.device_posture_attributes.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.
Delete a user?
intent: the user asked me to delete a user
account { op: "keys.create", name: "tailscale-agent", allowed_actions: ["tailscale.*"], cap_calls: 2000, confirm_policy: "human_for_writes", approval_rules: [ { match: "tailscale.users.delete", approver: "[email protected]" }, { match: "tailscale.keys.delete", approver: "[email protected]" }, { match: "tailscale.devices.delete", approver: "[email protected]" } ]}Cost
What Tailscale costs on actionlake.
Every Tailscale action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Tailscale 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 93 Tailscale actions
- 93actions pageeach page of a long list is one call
- 35reads can be cacheda cache hit costs nothing
- 100,000calls a month, freepaid plans from US$49 a month for 250,000
Action index
Every Tailscale action.
93 actions by area: the id your agent calls, what it does, and its effect.
Devices9
tailscale.devices.authorizeAuthorize devicewritetailscale.devices.deleteDelete a devicedestructivetailscale.devices.expire_keyExpire a device's keydestructivetailscale.devices.getGet a devicereadtailscale.devices.listList tailnet devicesreadtailscale.devices.set_ipSet device IPv4 addresswritetailscale.devices.set_nameSet device namewritetailscale.devices.set_tagsSet device tagswritetailscale.devices.update_keyUpdate device keywrite
Users7
tailscale.users.approveApprove a userwritetailscale.users.deleteDelete a userdestructivetailscale.users.getGet a userreadtailscale.users.listList usersreadtailscale.users.restoreRestore a userwritetailscale.users.suspendSuspend a userdestructivetailscale.users.update_roleUpdate user rolewrite
Webhooks7
tailscale.webhooks.createCreate a webhookwritetailscale.webhooks.deleteDelete webhookdestructivetailscale.webhooks.getGet webhookreadtailscale.webhooks.listList webhooksreadtailscale.webhooks.rotate_secretRotate webhook secretdestructivetailscale.webhooks.testTest a webhookwritetailscale.webhooks.updateUpdate webhookwrite
Device invites6
tailscale.device_invites.acceptAccept a device invitewritetailscale.device_invites.createCreate device inviteswritetailscale.device_invites.deleteDelete a device invitedestructivetailscale.device_invites.getGet a device invitereadtailscale.device_invites.listList device invitesreadtailscale.device_invites.resendResend a device invitewrite
Keys5
tailscale.keys.createCreate an auth key or trust credentialwritetailscale.keys.deleteDelete keydestructivetailscale.keys.getGet keyreadtailscale.keys.listList tailnet keysreadtailscale.keys.updateSet keywrite
Oauth apps5
tailscale.oauth_apps.createCreate an OAuth appwritetailscale.oauth_apps.deleteDelete OAuth appdestructivetailscale.oauth_apps.getGet OAuth appreadtailscale.oauth_apps.listList OAuth appsreadtailscale.oauth_apps.updateUpdate OAuth appwrite
Posture integrations5
tailscale.posture_integrations.createCreate a posture integrationwritetailscale.posture_integrations.deleteDelete a posture integrationdestructivetailscale.posture_integrations.getGet a posture integrationreadtailscale.posture_integrations.listList all posture integrationsreadtailscale.posture_integrations.updateUpdate a posture integrationwrite
User invites5
tailscale.user_invites.createCreate user inviteswritetailscale.user_invites.deleteDelete a user invitedestructivetailscale.user_invites.getGet a user invitereadtailscale.user_invites.listList user invitesreadtailscale.user_invites.resendResend a user invitewrite
Device posture attributes4
tailscale.device_posture_attributes.batch_updateBatch update custom device posture attributeswritetailscale.device_posture_attributes.deleteDelete custom device posture attributesdestructivetailscale.device_posture_attributes.getGet device posture attributesreadtailscale.device_posture_attributes.setSet custom device posture attributeswrite
Log streaming4
tailscale.log_streaming.disableDisable log streamingdestructivetailscale.log_streaming.getGet log streaming configurationreadtailscale.log_streaming.get_statusGet log streaming statusreadtailscale.log_streaming.setSet log streaming configurationwrite
Policy file4
tailscale.policy_file.getGet policy filereadtailscale.policy_file.previewPreview rule matchesreadtailscale.policy_file.setSet policy filewritetailscale.policy_file.validateValidate and test policy fileread
Services4
tailscale.services.deleteDelete a Servicedestructivetailscale.services.getGet a Servicereadtailscale.services.listList all Servicesreadtailscale.services.upsertUpdate a Servicewrite
Contacts3
tailscale.contacts.getGet contactsreadtailscale.contacts.resend_verificationResend verification emailwritetailscale.contacts.updateUpdate contactwrite
Split DNS3
tailscale.split_dns.getGet split DNSreadtailscale.split_dns.setSet split DNSwritetailscale.split_dns.updateUpdate split DNSwrite
Tailnets3
tailscale.tailnets.createCreate a tailnetwritetailscale.tailnets.deleteDelete a tailnetdestructivetailscale.tailnets.listList tailnetsread
Aws external ids2
tailscale.aws_external_ids.createCreate or get AWS external idwritetailscale.aws_external_ids.validateValidate external ID integration with IAM role trust policyread
Device routes2
tailscale.device_routes.getList device routesreadtailscale.device_routes.setSet device routeswrite
DNS configuration2
tailscale.dns_configuration.getGet DNS configurationreadtailscale.dns_configuration.setSet DNS configurationwrite
DNS nameservers2
tailscale.dns_nameservers.listList DNS nameserversreadtailscale.dns_nameservers.setSet DNS nameserverswrite
DNS preferences2
tailscale.dns_preferences.getGet DNS preferencesreadtailscale.dns_preferences.setSet DNS preferenceswrite
DNS search paths2
tailscale.dns_search_paths.listList DNS search pathsreadtailscale.dns_search_paths.setSet DNS search pathswrite
Service device approvals2
tailscale.service_device_approvals.getGet approval status of Service on a devicereadtailscale.service_device_approvals.updateUpdate approval status of Service on a devicewrite
Tailnet settings2
tailscale.tailnet_settings.getGet tailnet settingsreadtailscale.tailnet_settings.updateUpdate tailnet settingswrite
Configuration audit logs1
tailscale.configuration_audit_logs.listList configuration audit logsread
Network flow logs1
tailscale.network_flow_logs.listList network flow logsread
Service hosts1
tailscale.service_hosts.listList devices hosting a Serviceread
Questions
The Tailscale MCP server, answered.
Is there a Tailscale MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 93 Tailscale actions across devices, users, webhooks and device invites, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Tailscale.
What do I need to connect Tailscale?
Tailscale API access token (tskey-api-...): admin console > Settings > Keys > Generate access token. The free call account { op: "connections.requirements", service: "tailscale" } lists exactly what to paste.
Can an AI agent delete or change things in Tailscale?
Only if its key allows it, and only with an explicit confirmation. 15 Tailscale actions are destructive, including delete a user, delete key and delete a device. 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 Tailscale API limit?
actionlake paces every call to the limits in the Tailscale service file: 300 calls a minute per connection. Repeated reads are answered from cache (35 Tailscale actions can be), which never reaches Tailscale.
What does a Tailscale action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (93 Tailscale actions page). Cache hits, searches, dry runs and calls Tailscale 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 Tailscale API does it cover?
The Tailscale API v2 (OpenAPI 3.1) at api.tailscale.com/api/v2. It is built from Tailscale's documentation as read on 23 October 2026. 93 actions in 26 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Tailscale 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 Tailscale 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 Tailscale today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.