Provides read-only tools to query network health, device and active-client information from a Ubiquiti UniFi Network appliance.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@unifi-safe-mcp check network health and current devices"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A focused, read-only UniFi Network MCP server providing bounded health, device, and active-client views. Extracted from spectrenet-mcp commit 2e63f009d8f4642e17398685baef0251811d6637; this repository is self-contained.
Exactly three read tools: network_health, devices, and clients; no configuration, command, adoption, restart, block, or other mutation tools.
Appliance data requests use HTTP GET only (authentication necessarily uses POST /api/auth/login).
TLS certificate and hostname verification are enabled by default.
Authentication and appliance data redirects are rejected rather than followed.
The configured site is encoded as one URL path segment, and endpoint suffixes are closed-allowlisted.
HTTP response bytes, returned list lengths, projected fields, and audit fields are bounded.
Credentials and full responses are not audited.
Related MCP server: UniFi MCP Server
This targets the UniFi OS console proxy paths /api/auth/login and /proxy/network/api/s/{site}/..., observed on modern UniFi OS appliances. Older self-hosted UniFi Network Controllers commonly use /api/login and non-proxied /api/s/...; those are not supported. Ubiquiti changes private/internal endpoints between appliance and Network Application versions. No version matrix is certified: test this alpha against your exact console firmware and Network Application version in non-production, and pin upgrades operationally.
uv sync --extra test
uv run unifi-safe-mcp --help
uv run unifi-safe-mcpThe server uses MCP stdio and does not load .env. Treat .env.example as fake documentation only; inject a dedicated read-only account through a supervisor/secret manager.
uv sync --locked --extra test
uv run --locked pytest
uv run --locked python -m compileall -q src tests
uv buildSee architecture, compatibility, and security. MIT © 2026 Ben Gauger.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0clientsdevicesnetwork_healthScored across 3 tools
The three tools—network_health, devices, and clients—each address a distinct aspect of a UniFi network with no overlap in purpose. An agent can easily differentiate between them based on their names.
All three tool names are single nouns following a consistent pattern (no verbs or mixed conventions). While this differs from the common verb_noun style, the naming is internally consistent and predictable.
With only three tools, the server is minimal but within the acceptable range for a focused network management utility. The count feels proportionate to the narrow scope of health, device, and client visibility.
The server covers only basic read‑only operations for network health, devices, and clients. It lacks essential UniFi features such as site management, configuration changes, or traffic monitoring, leaving significant gaps for most real‑world use cases.
Read-only MCP access to a documented IT fleet: state, changes, posture. 15 tools.
Governed MCP gateway: one endpoint for your tools, with credential custody and audit log.
XFA's remote MCP server — query device posture, compliance, policies & CVEs. Read-only.
Read-only MCP access to sessions, funnels, campaigns, errors, live visitors, and anomalies.
