# Autonomous Vision Monitor (Claude)

> An unattended AI that watches, assesses, and only speaks up when something needs doing.

**Status:** Prototype

## Problem

Continuous monitoring either needs a human watching or brittle threshold rules that miss judgment calls.

## What it does

A domain-agnostic monitoring agent that runs entirely on its own: on a schedule it captures an image, pulls in supporting context (weather, sensor readings, recent history), and has Claude assess the scene against a per-domain rubric using vision, then logs the assessment and alerts a human only when action is actually needed. The architecture is a reusable kernel plus swappable domain packs, so the same agent monitors a garden or a fleet of services with no code change, just a new rubric. It runs headless (no server, no daemon) on cheap edge hardware.

## Measured results

- **Unattended** no human in the loop — Runs daily on a schedule and alerts only when action is needed.
- **1 kernel** many domains — Same agent monitors a garden or service health via a swappable rubric, no code change.

## Stack

Claude (vision, headless), Claude Code, Cron scheduling, Bash kernel + domain packs, Edge device (Raspberry Pi)

## Topics

AI, Agents, LLM, Architecture

---

Part of the proof-of-work portfolio of Michael Wahl — VP, Software Engineering & AI Enablement.

- Canonical project page: https://connect.michaelwahl.org/project/advisory-monitor-agent
- Full professional profile: https://connect.michaelwahl.org/
- Machine-readable context: https://connect.michaelwahl.org/llms.txt