bugsweep

Solid

系统化 Bug 扫荡工作流,包含基于代码库规模的目标值计算、翻倍升级机制、区域追踪及最终验证。适用于执行 /bugsweep 或用户要求进行系统化 Bug 排查时。

AI & Automation 4 stars 0 forks Updated today MIT

Install

View on GitHub

Quality Score: 83/100

Stars 20%
23
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

<img src="banner.png" width="100%" alt="bugsweep banner"> > **中文** — `bugsweep` 官方中文版本。 # /bugsweep — 系统化 Bug 扫荡工作流 (中文) 具有收敛停止条件的迭代式 Bug 排查流程。根据代码库规模按比例调整目标,在搜索显得过于表面时触发升级,并通过区域追踪防止重复检查。 ## 1. 计算基础搜索轮数 (base_rate) ``` LOC = productive source lines (src/, lib/ — excluding tests, configs, docs, generated) x = max(1, ceil(LOC / 1500)) base_rate = x * 3 ``` | LOC | x | 基础轮数 (Base rate) | |-----|---|-----------| | ~1500 | 1 | 3 | | ~3000 | 2 | 6 | | ~4500 | 3 | 9 | | ~10000 | 7 | 21 | 向用户汇报: "代码库规模: {LOC} LOC → 基础轮数 = {base_rate} 次无 Bug 搜索轮次。" ## 2. 搜索循环 ``` counter = 0 target = base_rate any_bug_found = False checked = [] # (area_name, type: code|task) LOOP: area = pick_new_area() # see area rules checked.append(area) Perform a thorough bug search IF bug found: any_bug_found = True Fix following bugfix-protocol (phases 4+5) Review: see model rule (newer model classes: no external review needed) Commit + push counter = 0 # RESET ELSE: counter += 1 Report: "✓ Clean: {area} — {counter}/{target}" IF counter >= target: IF NOT any_bug_found: # Doubling escalation: not a single bug → search too shallow? target = base_rate * 2 any_bug_found = True # escalate only ONCE Report: "⚠ No bug in {base_rate} passes → target doubled to {target}." CONTINUE LOOP ELSE: GOTO final verification ``` ### 搜索循环实践注意事项 (来自实际扫荡经验) - **非 Git 仓库:** 在没有 `git` 的环境中(例如云端同步的项目文件夹),使用**版本化备份**代替 "commit + push":...

Details

Author
ellmos-ai
Repository
ellmos-ai/skills
Created
6 months ago
Last Updated
today
Language
Python
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category