@yawlabs/tailscale-mcp
Related Servers
Alternatives to @yawlabs/tailscale-mcp
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityDmaintenanceQuery devices, check connectivity, and get network status from your Tailscale tailnet directly via an AI assistant.1MIT
- AlicenseAqualityCmaintenanceEnables AI agents to securely interact with Tailscale tailnets for device inventory, ACL review, key hygiene, and more, with token-efficient output and write-gated mutations.19MIT
- AlicenseNot gradedqualityDmaintenanceWraps the Tailscale CLI to enable MCP-capable AI agents to manage devices, serve configurations, and diagnose connectivity.3MIT
- FlicenseNot gradedqualityNot gradedmaintenanceEnables interaction with the Tailscale API to manage devices, ACLs, keys, and DNS configurations directly from MCP clients like Claude Desktop or Raycast.-
- AlicenseNot gradedqualityBmaintenanceEnables administering a Tailscale tailnet through a single action-router tool that maps requests to the Tailscale API v2: listing and inspecting devices and routes, reading keys, ACL policy, DNS settings and users, and authorizing devices. Device deletion is double-gated behind a confirm flag plus an environment opt-in, refusing the call before any request if either gate is missing.66 npmMIT
- AlicenseAqualityCmaintenanceEnables Claude to connect and disconnect the host machine from its Tailscale network via the local CLI, and to manage the wider tailnet (devices, routes, DNS, ACL policy, auth keys, webhooks, users, audit log) through the Tailscale REST API. Access is governed by configurable risk tiers, with destructive operations gated behind interactive human approval.1840 npmMIT
TDQS
Scored across 98 tools
Most tools target clearly distinct resources and actions, but several clusters overlap heavily: DNS can be changed through set_dns_preferences, set_nameservers, set_search_paths, set_split_dns, update_split_dns, and set_dns_configuration; device authorization has both single-device and batch variants; and key/OAuth-app/token operations require careful reading to separate. The descriptions do help, but with 98 tools an agent can still easily choose the wrong DNS or authorization entry point.
Names follow a highly predictable tailscale_verb_noun pattern in snake_case, such as tailscale_list_devices, tailscale_get_device, tailscale_create_key, and tailscale_delete_webhook. The few exceptions like tailscale_status and tailscale_tool_groups are clearly intentional utility tools and do not undermine the convention.
Ninety-eight tools is an extreme surface for a single MCP server and far exceeds what an agent can efficiently discover, select, and safely operate. Even though each tool maps to some Tailscale API capability, the sheer count creates high selection cost and makes the toolset unwieldy without strong grouping or progressive disclosure.
Coverage is broad across devices, users, keys, DNS, ACLs, webhooks, invites, posture, logging, and organization tailnets, with lifecycle operations mostly present. The notable gap is service creation, which the server explicitly says it does not support, plus some DNS operations are split across overlapping endpoints, so agents may need workarounds in a few areas.