Top 200 ReposEXPLORER
Browse agents, categories and pages
Back to results
Planned, not doneThe source describes an intention, not a finished result.

Turn a camera snapshot into a sensor with Home Assistant’s AI Task

Home Assistant 2025.8 added AI Task: an automation sends a camera image to your chosen AI with a question, gets a structured answer back and stores it, for example as a count of birds in a coop.

Try the promptRead the original (opens in a new tab)

The problem

How it works

Home Assistant’s AI Task integration lets automations, scripts and template entities hand a job to an AI model and get the result back as text or as fields you define. You add an AI Task entry under an AI integration such as Ollama or OpenAI, then call ai_task.generate_data with instructions, an optional output structure and attachments such as a camera image. The documented example is a template sensor that runs every five minutes, attaches a snapshot from a coop camera, asks how many birds are inside and stores the number as a sensor. The docs suggest the same pattern for free parking spaces, people in a room or whether a door is open, and the release added a blueprint for analysing camera footage.

Results and limitations

The source is official documentation and release notes with example configurations, not a measured deployment, so this case is marked planned.

Before you try it

Each run sends a camera image to whichever model you configure; use a local model through Ollama if images should stay at home, and think about who appears on camera. A five-minute schedule means many model calls per day, which costs money with a cloud API. AI features stay hidden unless you set up an AI integration. Treat counts as estimates rather than alarms you rely on for safety. Results depend on your setup and were not independently reproduced by Top 200 Repos.

Tools and repositories

Tools
Home Assistant 2025.8 or later · An AI integration such as Ollama or OpenAI · A camera entity
Prerequisites
A Home Assistant installation with a camera, plus a local model or a cloud AI API key.

Try it yourself: Write the question an AI Task asks about a camera

Fill the fields to adapt the prompt. Empty fields stay as visible [placeholders].

Preview
This is a snapshot from my [Camera location] camera. Answer only from what is visible in this image. How many [What to count] can you see? If the view is blocked, too dark or unclear, return [Value when unclear] instead of guessing. Return the number in the field [Output field name].

Use this as the instructions of ai_task.generate_data with a number field in the structure, and test it on daytime and night-time snapshots first.

Related videos

About ollama/ollama

Independent creators. Listing a video implies no endorsement, sponsorship or partnership.

Sources

  1. AI Task (opens in a new tab)

    Project documentation · home-assistant.io · Undated · Source checked 12 Nov 2026, 09:55 UTC

    What this source supports
    • AI Task is a building block used by AI integrations, with a preferred AI Task entity per task
    • Actions include ai_task.generate_data and ai_task.generate_image
    • Example template sensor attaches a coop camera image every five minutes and stores the bird count as a number
    • Suggested alternatives: free parking spots, people in a room, whether a door is open
  2. 2025.8: The summer of AI (opens in a new tab)

    Project documentation · home-assistant.io · Published 6 Aug 2025 · Source checked 12 Nov 2026, 10:08 UTC

    What this source supports
    • AI Task was introduced in Home Assistant 2025.8 and can take file and camera attachments
    • Output can be text or a structure of your choice, from automations, scripts and template entities
    • Ollama servers and OpenAI API keys are configured once and used through sub-entries
    • AI features do not appear unless an AI integration is set up, and a blueprint for analysing camera footage was provided
Report a problem

Reported by source

Control smart-home devices with a local LLM voice assistant

Connect Home Assistant to a local Ollama model as its conversation agent so you can ask it to control exposed devices privately.

github.comhome-assistant/core, ollama/ollama
Planned, not done

Auto-tag receipts and bills in Paperless-ngx with paperless-ai

paperless-ai suggests titles, tags, correspondents and dates for documents in a Paperless-ngx archive, using a local Ollama model or a cloud LLM, with manual review and undo.

github.comollama/ollama
Reported by source

Extract invoice fields locally with a vision model on Ollama

Stephen Sebastian replaced a paid cloud OCR API with Gemma 4 E4B on Ollama to pull date, amount and description from contractor invoice photos on his laptop.

dev.toollama/ollama
Planned, not done

Record and summarise meetings on your own machine with Meetily

Meetily records calls, transcribes them on the device with Whisper or Parakeet, and writes summaries with a local Ollama model or a cloud model you pick.

github.comollama/ollama
Reported by source

Run an open-weight model locally and call it from your code

Install Ollama, pull an open model and call it through the local REST API so prompts stay on your machine.

github.comollama/ollama
Planned, not done

Run OpenAI’s Codex CLI on open models served by Ollama

Ollama’s docs show how to point Codex CLI at local tool-capable models such as gpt-oss with one launch command or a saved profile, switch models inside a session and turn off web search.

docs.ollama.comopenai/codex, ollama/ollama