PROTOTYPE · HUMAN-IN-THE-LOOP AGENTS

Automation that knows when to ask permission.

HomeOps Agent explores explicit boundaries for useful household automation.

Problem

Household automation needs to distinguish a safe suggestion from an action that requires a person’s approval.

Engineering question

How can an agent use tools while keeping permission, failure, and audit behavior explicit?

What I built

A prototype with agent tools, tool contracts, approval gates, deterministic scenarios, and failure handling.

Architecture

The agent selects bounded tools. Approval gates interrupt actions that require human authorization.

Evaluation

Deterministic scenarios exercise expected behavior, including permission and failure paths.

Guardrails and boundaries

Current device integrations are simulated for reproducible testing. Human approval remains part of the design.

Failure evidence

Failure handling is modeled as a first-class path; this prototype does not establish reliability in a real household.

Tradeoffs

Simulation makes behavior repeatable but does not validate real device behavior or operational resilience.

Current limitations

This is a prototype, not a production deployment or a real device integration.

What comes next

Extend validation before connecting any real device or durable action.

View repository ↗