Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Home Assistant Automation Debugger

    1

    Find out why your Home Assistant automation misbehaved and what to check next.

    Free

    1 useSecurity scanned
    Home Assistant Automation Debugger

    Home Assistant Automation Debugger

    Example session with this skill installed

    My Home Assistant hallway light stayed off. Trace: motion trigger fired; top-level condition sun.sun = below_horizon was false; sun.sun was above_horizon. Why?

    • Read your context and instructions
    • Compiled the home assistant automation

    The automation triggered, but its sun condition stopped the run before the light action. The evidence points to a condition blocking the action, not a missed motion trigger.

    Diagnosis

    Evidence you suppliedWhat it means
    Motion trigger firedThis run started from the motion trigger.
    Top-level condition required sun.sun to be below_horizonThe automation required the sun to be below the horizon before continuing.
    Actual state was above_horizon; condition was falseThat requirement was not met, so the run stopped before its actions.

    Explanation: The condition explains why this run did not turn on the light. If the light should operate only after sunset, this result matches that requirement. If you wanted it to work during the day too, the condition may not match your intent.

    Still unknown: Whether daytime lighting should be allowed. This trace also does not establish whether the light action itself works, because it was not reached.

    Next steps

    • Inspect the sun condition in the automation editor without running the actions.
    • Tell me whether the light should work during daylight before we consider a change.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    A light stays off, an automation runs twice, or a routine stops halfway through. Get an evidence-based explanation of what went wrong and a focused next check, without giving anyone access to your home.

    What it does

    • Finds the stopping point: Reviews triggers, conditions, branches, and actions to explain where a run changed course.
    • Investigates unexpected behavior: Examines missed runs, repeated actions, timing issues, and differences between manual and automatic runs.
    • Separates facts from guesses: Shows what your evidence supports and what still needs checking.
    • Suggests targeted changes: When requested, proposes a small YAML change and explains its effect.

    How it works

    1. Describe what you expected and what happened instead.
    2. Share relevant configuration, traces, screenshots, or logs with sensitive details removed.
    3. Receive a diagnosis, any important uncertainty, and a low-risk next check.
    4. Ask for a proposed change if needed, then review it before applying it yourself.

    Works with

    Home Assistant automation YAML, UI configuration, exported traces, entity states, screenshots, and logs. Basic analysis does not require internet access.

    Why use this skill

    Avoid changing triggers when a condition or action is the actual problem. Focus your troubleshooting on the supplied evidence instead of working through unrelated advice.

    Use cases

    • Explain why a motion-triggered light stayed off.
    • Investigate repeated runs or skipped branches.
    • Examine failures after an entity rename or restart.

    Known limitations

    It cannot connect to Home Assistant or control devices. Conclusions depend on the evidence you share, and version-specific behavior may need verification.

    Analyzes only supplied evidence. Cannot connect to Home Assistant, control devices, or apply fixes. Version-specific behavior may need verification.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed1 month ago
    Updatedtoday

    What's inside

    Frequently Asked Questions

    Simply provide a description of the issue along with your automation YAML, trace JSON, or logs; the skill will analyze the logic and state data to pinpoint exactly where the sequence failed or diverged.

    No, this skill operates as a standalone logic analyzer to ensure your privacy; it identifies issues based solely on the data you choose to paste or upload without needing remote access to your server.

    The skill is compatible with both YAML-based configurations and UI-created automations, as long as you can provide the configuration text or an exported trace file.

    Unlike generic AI, this skill is specifically tuned to recognize Home Assistant’s unique trigger logic, condition processing, and trace formatting to provide evidence-based fixes rather than generic advice.

    You receive a specialized diagnostic framework that interprets complex trace logs and provides targeted YAML recommendations to resolve common issues like 'flapping' triggers or failed conditions.

    New in Testing & Debugging