Reolink MCP logo

Reolink MCP

Connects AI clients to local Reolink cameras for snapshots, monitoring, PTZ, lighting, siren, and other device controls.

Install Reolink MCP

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "reolink-mcp": {
      "command": "uvx",
      "args": [
        "reolink-mcp"
      ],
      "env": {
        "RMCP_CONFIG_FILE": "YOUR_VALUE_HERE",
        "RMCP_CAMERAS__NAME__PASSWORD": "YOUR_VALUE_HERE",
        "RMCP_READ_ONLY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

uvx reolink-mcp

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 1d ago.

Overview

The reolink-mcp MCP server connects MCP clients directly to Reolink cameras on a local network. It provides tools for snapshots, device details, capabilities, motion and AI detection states, PTZ presets, zoom, lighting, and siren controls. Configuration uses YAML for camera hosts and environment variables for passwords. Use it when an agent needs direct camera access without an NVR or home-automation service.

Use cases

  • View live snapshots from configured Reolink cameras
  • Inspect camera state and AI detection results
  • Control sirens, spotlights, IR lights, and white LEDs
  • Move PTZ cameras to saved presets
  • Run camera monitoring in read-only mode

Key features

  • Local-network Reolink camera access
  • MCP image snapshots
  • Device, capability, motion, and AI-state inspection
  • Siren, lighting, zoom, and PTZ controls
  • Capability checks for control operations
  • Read-only startup mode

Tools16

What Reolink MCP lets your agent do. Expand a tool for its parameters.

Self-reported~362 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 1d ago

Recent health check succeeded.

6/6 checks healthy over the last 41d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 5d ago via OSV.dev · reolink-mcp (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars3Stargazers on the source repository.
Last commit3mo agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Tools exposed16Callable tools this server registers over MCP.

How Reolink MCP works

What reolink-mcp MCP server does

The reolink-mcp MCP server exposes local Reolink camera operations through the Model Context Protocol. It can enumerate configured cameras, report connection status and hardware information, inspect supported features, and retrieve current device states. Detection polling includes plain motion plus camera-reported AI categories such as person, vehicle, and pet.

The server also returns live still images through MCP image content blocks. For cameras with the relevant hardware, agents can control the siren, audio-alarm setting, spotlight, IR lights, white LED, zoom, and PTZ functions. PTZ operations include listing presets, moving to a preset, reading the current position, and configuring a guard point with automatic return behavior.

How it works

Communication goes directly from the server to cameras on the local network. No cloud service, NVR, Home Assistant instance, or other intermediary is described as required. The implementation uses the reolink-aio library and the official MCP Python SDK, and maintains one long-lived session for each configured camera while the process runs.

Tools are divided into observation and control operations. Observation tools read information without changing the camera. Control tools can mutate camera state, while set_siren is specifically marked destructive. MCP clients can use the tool annotations to manage approval prompts.

Control requests are checked against the camera's reported capabilities before being sent. Unsupported functions should produce a clear refusal rather than a raw device API error. Siren activation defaults to about five seconds and rejects durations above 60 seconds instead of reducing them automatically.

Setup and configuration

Install the server with a generic stdio MCP configuration, commonly using uvx reolink-mcp. Create a YAML file at ~/.config/reolink-mcp/config.yaml, or provide another path through RMCP_CONFIG_FILE. Each camera entry needs a key, host, and username. Passwords are supplied separately through an environment variable named RMCP_CAMERAS__<name>__PASSWORD, where the name matches the YAML key exactly.

Claude Code and Claude Desktop configuration examples are provided in the project material. Other MCP clients can use the same command and arguments through stdio. RMCP_READ_ONLY=true removes all control tools when the server starts, leaving only observation tools registered.

Limitations and notes

The reolink-mcp MCP server requires a reachable Reolink camera and valid administrator credentials. Camera session capacity is a practical deployment constraint: the server keeps one session per configured camera, so another NVR, app, automation system, or MCP process may cause login or session errors if the camera's limit is exceeded.

Live development validation covered siren, spotlight, and zoom on a P437, and IR lights on a P320. PTZ operations were mock-validated only, so real PTZ hardware validation remains pending. Feature availability also depends on each camera's capabilities; the server checks those capabilities before issuing control commands.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Reolink MCP

reolink-mcp is an MCP server that connects AI clients directly to Reolink cameras on a local network. Its main tools retrieve snapshots, device state, capabilities, motion and AI detection results, while other tools control PTZ, zoom, sirens, spotlights, IR lights, and white LEDs without requiring an NVR or cloud service.

Install and run it through an MCP client's stdio configuration with `uvx reolink-mcp`. Create the YAML camera configuration first, then provide each camera password through its matching `RMCP_CAMERAS__<name>__PASSWORD` environment variable.

Set one `RMCP_CAMERAS__<name>__PASSWORD` environment variable per configured camera. The `<name>` portion must exactly match the camera key in the YAML file; passwords should not be placed in that file.

Yes. Claude Desktop can run it with `uvx reolink-mcp` in the `mcpServers` configuration. Pass the camera password through the configuration's `env` object and ensure the camera is reachable on the local network.

Set `RMCP_READ_ONLY=true` when starting the server. It removes all control tools at startup, so the MCP client receives an observe-only tool registry.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewReolink MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/reolink-mcp?style=directory)](https://allmcps.com/mcp/reolink-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/reolink-mcp"><img src="https://allmcps.com/api/badge/reolink-mcp?style=directory" alt="Reolink MCP on AllMCPs" /></a>