ac-debug

Coordinate main-thread diagnosis with subagents to collect evidence and plan fixes.

6|2|Updated Jan 9, 2026
One-click install
npx skills add https://github.com/yz0812/ai-dev-toolkit --skill ac-debug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ac-debug
Source: https://github.com/yz0812/ai-dev-toolkit/tree/main/skills/ac-debug
Command: npx skills add https://github.com/yz0812/ai-dev-toolkit --skill ac-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

问题诊断:定位报错与异常根因,按需并行取证,并在用户确认后执行最小修复与验证。

Core Features & Use Cases

    1. 主线程与 subagents 协同诊断与取证,输出可执行的修复建议。
    1. 支持跨模块的调用链分析、日志聚合与证据整理,确保可复现性与可追溯性。
    1. Use Case: 当同一问题出现在多个模块时,自动分发线索到相关子代理并整合证据。

Quick Start

仅需将问题描述输入到 /ac-debug 命令,系统将启动诊断流程并在获得用户确认后给出修复方案。

Frequently Asked Questions about ac-debug

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

FAQPage Schema
How do I diagnose errors across multiple modules using trace analysis?

Multi-module trace analysis diagnostics coordinate a main thread with subagents to gather evidence across components, aggregating logs to pinpoint root causes. This staged workflow ensures cross-component traces are captured and validated before any repair is applied.

What is the best way to aggregate logs for debugging complex failures?

Log aggregation for debugging failures is handled by dispatching clues to subagents that collect and integrate evidence across affected modules. This parallel evidence gathering ensures full traceability and reproducibility for complex error diagnosis.

How do I start a structured diagnostic workflow for cross-component errors?

To start a structured diagnostic workflow for cross-component errors, input the issue description into the command interface. The system initiates staged evidence gathering, coordinates subagents for trace analysis, and presents a user-validated repair plan.

Does this debugging approach require user confirmation before applying fixes?

Yes, this debugging approach requires explicit user confirmation before applying any fix. It generates actionable repair suggestions only after the main thread and subagents complete evidence collection and validate the root cause analysis.

Can I use subagents to diagnose issues when the same error appears in multiple modules?

You can use subagents to diagnose issues when the same error appears in multiple modules by automatically dispatching clues to relevant subagents. They gather evidence in parallel and integrate findings to ensure comprehensive cross-component trace coverage.

When should I not use a coordinated subagent workflow for issue diagnosis?

You should not use a coordinated subagent workflow for issue diagnosis when dealing with single-module errors that lack cross-component traces, as the overhead of dispatching and integrating subagent evidence outweighs the benefits of parallel collection.