More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Automation Brittleness Fix

    2

    Find why an automation breaks on small changes and harden it step by step.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Automation Brittleness Fix

    Automation Brittleness Fix

    Example session with this skill installed

    Apply the skill to this case: A nightly report automation failed about twice a week. Each time someone fixed a column position or a wait time, and it broke again differently the next week.

    • Read your context and instructions
    • Compiled the automation brittleness fix

    Situation: A nightly report automation failed about twice a week. Each time someone fixed a column position or a wait time, and it broke again differently the next week. Evidence: Of the last 20 failures, nine were timing, six were column position, and five were file-format changes. The script used fixed sleeps and read columns by position, so any upstream change broke it. Decision: Harden in cause order: condition waits first, then named-column reads, then a format check that rejects unknown formats with a clear message instead of guessing. Outcome: Zero failures in the next 30 days, and the one later upstream change produced a clear format error within minutes instead of a silent wrong report.

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

    About this skill

    Brittle automations do not fail because the task is hard. They fail because they depend on exact text, exact timing, exact screen position, or an undocumented assumption that changed. Each fix patches one break and leaves the next one waiting. This skill inventories every fragile dependency, ranks them by break frequency, and replaces the worst ones with tolerant checks, retries, and fallbacks. This skill gives you a counted diagnosis, a decision table, a fill-in template, a checklist, and a scoring rubric, plus a runnable script (brittleness_scorer.py) that Scores a failures CSV by cause, ranks causes with each cause's share of failures, and writes a hardening report with thresholds. It is built for the bottleneck moment: when the usual quick fix has already failed, or when a silent failure is worse than a loud one. Follow the protocol in order, record numbers rather than impressions, and use the quality gate before you call the work done.

    Results depend on the quality of your evidence set: an incomplete failure log, missing dates, or causes recorded inconsistently will limit the diagnosis, and the skill says so rather than filling gaps with guesses. The script is a helper for counting and reporting; it does not replace judgement about safety, permission, or whether a fix is correct. Thresholds are starting points to adapt to your context, not universal constants.

    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

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 23 days ago

    Permissions

    Terminal / Shell
    Write Files
    Listed23 days ago
    Updated21 days ago

    What's inside

    Frequently Asked Questions

    Only if the failure count shows the design itself is wrong, such as depending on screen positions for a data task. Most brittle automations improve faster by hardening the top two causes.

    Retries are fine when the action is safe to repeat and capped. They are harmful when they hide a cause or repeat a paid or destructive action.

    Re-run the exact failing inputs, plus good inputs, plus a peak-size input, and watch 30 days of runs before declaring it stable.

    New in Workflow Automation