subtask-update

Diagnose gate-check failures and re-enter the autoworker loop.

17|3|Updated Mar 24, 2026
One-click install
npx skills add https://github.com/phj128/autoworker --skill subtask-update
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subtask-update
Source: https://github.com/phj128/autoworker/tree/main/skills/subtask-update
Command: npx skills add https://github.com/phj128/autoworker --skill subtask-update

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps teams quickly adapt a failed subtask plan by auto-reading gate-check results, diagnosing issues, proposing targeted corrective steps, and re-entering the autoworker loop to resume progress.

Core Features & Use Cases

  • Auto-detects gate-check failures and extracts failure details.
  • Designs mode-A supplementary steps and mode-B user feedback fixes.
  • Updates subtask.md and triggers autoworker:dispatch to continue flow.
  • Use Case: after a gate-check FAIL, auto-suggests fixes and re-dispatches to re-run checks.

Quick Start

Trigger the auto-update by providing a gate-check failure summary or a user bug description to re-enter the autoworker:dispatch loop.

Frequently Asked Questions about subtask-update

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate fixing failed subtasks after a gate-check failure?

Failed subtasks are automatically refined by diagnosing gate-check failures, generating targeted corrective steps with explicit scripts, and re-dispatching into the autoworker loop to resume processing.

What is the autoworker loop's process for handling user-reported bugs during subtask testing?

User-reported bugs trigger mode-B feedback fixes where the system reads failure details, designs supplementary corrective steps, updates subtask.md, and invokes autoworker:dispatch to continue the workflow.

How to re-dispatch a corrected subtask workflow step by step?

Provide a gate-check failure summary or bug description, allow the system to generate mode-A supplementary steps, update subtask.md with explicit actions, then trigger autoworker:dispatch to re-enter the execution cycle.

Does subtask automation require any dependencies to re-enter the execution cycle?

No external dependencies are required; the diagnostic and re-dispatch process operates independently by parsing gate-check outputs and immediately invoking the autoworker:dispatch loop to continue processing.

Why does my automated subtask workflow keep failing after re-dispatch?

Persistent failures occur when generated corrective steps lack explicit scripts or actions; the system requires precise analysis of gate-check outputs to design effective mode-A supplementary steps before re-entering the autoworker loop.

Can I use this subtask correction approach for both automated gate-check failures and manual user bug reports?

Both scenarios are supported: mode-A supplementary steps address automated gate-check failures, while mode-B fixes handle user-reported bugs, with both paths updating subtask.md and triggering autoworker:dispatch.