Control QVR Client with AI Agents (Beta)

At Computex 2026 and QNAP World Tour 2026, QNAP demonstrated AI agents operating QVR Client. The agent receives a task in natural language, then switches cameras, searches recordings, exports clips and reports the results.

Current status: Release Testing. Supported AI agents include ChatGPT Codex, Claude Code and Antigravity CLI. A beta build is available. The request procedure is described at the end of this post.

Supported tasks

Input Agent action
“Show the Lobby camera full screen.” Resolves the camera name. Loads the camera.
“When did Alex appear yesterday afternoon?” Runs a face search for that time range.
“Was plate ABC-1234 seen this week?” Runs a license plate search. Summarizes the results.
“Export the last 5 minutes of the Door camera.” Exports the clip. Monitors progress until the file is saved.
“Create a rule: motion on the Walkway camera starts recording.” Creates the event rule. Confirms it was saved.

The current build provides 92 automation methods. Coverage: live view, playback, search, export, bookmarks, event rules, schedules, PTZ and screenshots. Multi-step requests are supported. Example: “Find every face match for Alex this morning, export a 30-second clip around each, and summarize them in a table.”

How it works

  1. Discovery. The agent queries QVR Agent, the registry service installed with the client, for running clients.
  2. Contract retrieval. The agent retrieves the session rules, the method list and each method’s parameters from the client.
  3. Execution and verification. The agent sends JSON-RPC commands over WebSocket. After each action, it checks the client state.

The client documents its own interface. Any agent that can open a WebSocket connection can use it. Methods added in later builds are detected without updating the skill.

Setup

  1. In QVR Client, open Settings > AI Agent Control and enable Accept AI Agent commands. The client starts a local-only endpoint. Disabling the option stops the service immediately.
  2. Under Skill Installation, copy the prompt and paste it into the AI agent’s chat (example: Claude Code). The agent installs the qvr_ctrl skill bundled with the client.

Setup is complete. Suggested first input: “Can you control QVR Client on my computer?”

Security and privacy

  • Activation. The service runs only while Accept AI Agent commands is enabled.
  • Scope. The endpoint listens on 127.0.0.1. It cannot be reached from the network.
  • Credentials. Passwords, PINs and camera credentials are not exposed to the agent.
  • Permissions. The agent operates within the permissions of the logged-in account.
  • Method labels. Each method is marked read-only or write. The agent can be limited to read-only methods, or each change can require approval.

Warning: Data the AI agent reads, including camera lists, server status, screenshots and recordings, is sent to the AI provider’s cloud under that provider’s privacy policy. Use with cameras that capture sensitive scenes is not recommended. Results vary by AI agent and model.

Requirements and limits

Requirements: the QVR Client beta on Windows or macOS, and an AI agent running on the same computer.

Limits in this beta:

  • Remote control is not supported.
  • Some methods are not yet available.
  • Sign-in must be performed by the user.

Get the beta

To request the installation files, leave a comment on this post or send me a private message. The installation package will be provided.