nubjs/nub✦ 精选0°

impact-analysis

仓库创建 2026年6月4日最近提交 2 小时前SkillHot 收录 1 小时前
▸ 精选理由

帮助审查者全面评估改动影响,减少隐蔽回归与兼容问题。

▸ 风险提示

静态分析可能不全,运行时兼容性需人工复核和测试。

这个 Skill 做什么

基于 diff 系统追踪变更符号的调用点与潜在影响范围,评估连锁破坏风险。

拿一段 diff 去追踪所有被改动符号会影响到的地方,找到所有调用点、字段读写、接口实现和序列化/跨进程的连锁影响。适合改动范围大、影响面广、或关乎安全与对外契约的变更,目标是在合并前把那些“看起来没问题却远端崩溃”的情况抓出来。特点是做跨文件、跨模块的传递性追踪,能把隐藏的破坏面(blast radius)可视化并提示风险。

▸ 展开 SKILL.md 英文原文

The impact-analysis reviewer lens. Given a diff, systematically trace each changed symbol's BLAST RADIUS through the whole codebase (all call sites of a modified function, all readers/writers of a changed field, all impls/match-arms of a changed trait/enum, downstream behavioral/serialized/cross-process effects) so a locally-correct change that breaks a distant caller is caught BEFORE merge. Invoke (via the Skill tool) when a change earns a dedicated fresh-context reviewer — wide blast radius, security posture, serialized format, public surface — or when asked to "trace the impact", "find the blast radius", "what does this change break", "who calls this", "impact analysis".

开发编程影响分析代码审查静态追踪通用
3.9k
Stars
54
Forks
38
仓库内 Skill
积累中
7 日增星
安装 / 使用
给你的 Agent 一句话(通用)
帮我安装这个 skill:https://raw.githubusercontent.com/nubjs/nub/main/.claude/skills/impact-analysis/SKILL.md
或 curl 直取 SKILL.md
curl -fsSL "https://raw.githubusercontent.com/nubjs/nub/main/.claude/skills/impact-analysis/SKILL.md"
SKILL.MD 节选查看完整文件 ↗
# Impact analysis

A self-review answers two different questions. The **correctness** lens asks *"is the changed code itself right?"* The **impact** lens asks *"what ELSE did this change just break — silently, at a distance?"* A change can be locally flawless and still break a caller three files away that relied on the old return contract, a serialized format that's now incompatible, or a match arm that no longer compiles.

**When to buy it:** AGENTS.md governs escalation — a dedicated fresh-context reviewer is an escalation for a change that earns it (wide blast radius, security posture, serialized format, memory/UB-adjacent), not a default leg on every diff, and it never substitutes for ru
via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有