技能更新

工程开发 社区

技能解读

当前原文的自动摘译,尚未人工审校。

硬上限(由 script/skilllinter.py 强制执行)

SKILL.md ≤ 150 行 reference/*.md ≤ 200 行(reference/scenarios/*.md ≤ 400 行) README.md ≤ 100 行 每个 SKILL.md 都有 YAML frontmatter 与 name + description 没有 DO NOT / MUST NOT / NEVER 在## Anti-Patterns 部分之外

原则

简洁第一。每个文件都简短、简单、可读。挑战每一个令牌。渐进式披露。 SKILL.md 导航; reference/ 包含细节; reference/scenarios/ 包含具体的漏洞利用流程。分离关注点。 SKILL.md = 内容 + 何时。 reference/role-*.md = 代理生成时的行为方式。适用于任何横切规则(输出规则、凭证加载、暴力破解等)的单一规范主页。其他文件参考,不再重述。

何时更新现有技能

处理已完成的交战中的技术和故障模式。仅当四个条件都满足时,才能将学习提升到技能基础: 可推广。可重复使用的模式,而不是针对特定目标的知识。没有机器名称、实验室 ID、目标 IP、保留的标志、写入属性。物质改进。为未来的合作增加覆盖范围、效率或决策质量。技能库其他地方尚未捕获。 (scripts/skill_linter.py 标记重复项。)

重构配方

始终将框架构建为可重复使用的模式:“当遇到 X 条件时,尝试 Y 方法” - 永远不会 “在盒子 N 上,Y 有效”。在工具示例中使用 <TARGET_IP>、<DC_FQDN>、<DOMAIN> 占位符。

预写检查

在写入之前,运行python3 scripts/skill_linter.py。拒绝任何以下更改: 重新引入特定于挑战的知识。将SKILL.md推过 150 或参考值超过其上限。复制单一所有者规则(暴力、输出规则、环境读取器)。在反模式块之外添加 DO NOT / MUST NOT / NEVER。

流狐档案 运行与权限由流狐静态检查推断,不代表作者声明或实际测试结果。
领域
工程开发
作者声明的智能体
未找到明确声明;不据此推断已兼容或已测试
静态检查
88 / 100 · 启发式扫描,不代表运行安全
运行环境证据
未执行;当前没有隔离运行测试证据
作者 / 版本 / 许可证
@transilienceai · 未声明许可证
上下文用量评级
低消耗
接入复杂程度
需简单配置
是否需要外部接口密钥
不需要
兼容的系统
未声明(默认跨平台)
运行环境要求
未检出明确要求
检测到的文件与系统行为
  • 只读
  • 允许写入 / 修改
  • 命令执行
网络行为
仅限本地
安装命令数
无(仅作为资料)

需要注意: 未声明工具权限范围;实际权限由运行环境决定。

作者说明文件(原文) SKILL.md

# Skill Update

Generate or refine Claude Code skills following Anthropic best practices.

## Hard caps (enforced by `scripts/skill_linter.py`)

- `SKILL.md` ≤ 150 lines
- `reference/*.md` ≤ 200 lines (`reference/scenarios/*.md` ≤ 400 lines)
- `README.md` ≤ 100 lines
- Every `SKILL.md` has YAML frontmatter with `name` + `description`
- No `DO NOT` / `MUST NOT` / `NEVER` outside an `## Anti-Patterns` section
- No challenge-specific identifiers (machine names, lab IDs, lab IPs, preserved flags)
- Every Markdown link resolves to an existing file
- Every reference file is linked from at least one other file (no orphans)

## Principles

- **Brevity first.** Every file short, simple, human-readable. Challenge every token.
- **Progressive disclosure.** SKILL.md navigates; `reference/` holds detail; `reference/scenarios/` holds concrete exploit flows.
- **Separation of concern.** SKILL.md = WHAT + when. `reference/role-*.md` = HOW agents behave when spawned.
- **Single canonical home** for any cross-cutting rule (output discipline, credential loading, brute-force, etc.). Other files reference, never restate.

## File structure

```
skills/<skill-name>/
├── SKILL.md           # ≤150 lines, YAML + navigation
├── reference/
│   ├── *-principles.md  # ≤150 lines (decision tree)
│   ├── INDEX.md
│   ├── *.md             # patterns, ≤200 lines
│   └── scenarios/
│       └── <category>/
│           └── *.md     # ≤400 lines, self-contained
└── README.md          # optional, ≤100 lines
```

## SKILL.md template

```yaml
---
name: <skill-name>
description: What it does AND when to use. Include trigger phrases.
---

# <Skill Name>

<one-paragraph scope>

## When to use

- <bullet>

## Workflow / Quick start

<≤30 lines>

## References

- [reference/...](reference/...)

## Anti-Patterns

- <when negative framing is genuinely needed, put it here>
```

## When to update an existing skill

Process the techniques and failure modes from completed engagements. Promote a learning to the skill base only if **all four** hold:

1. **Generalizable.** Reusable pattern, not target-specific lore. No machine names, lab IDs, target IPs, preserved flags, writeup attributions.
2. **Material improvement.** Adds coverage, efficiency, or decision-quality for future engagements.
3. **Not already captured** elsewhere in the skill base. (`scripts/skill_linter.py` flags duplicates.)
4. **Minimal footprint.** Prefer extending an existing entry over adding a new file. Keep the base lean and high-signal.

## Reframing recipe

Always frame as a reusable pattern: *"when encountering X condition, try Y approach"* — never *"on box-N, Y worked"*. Use `<TARGET_IP>`, `<DC_FQDN>`, `<DOMAIN>` placeholders in tool examples.

## Pre-write check

Before writing, run `python3 scripts/skill_linter.py`. Reject any change that:
- Re-introduces challenge-specific lore.
- Pushes a `SKILL.md` past 150 or a reference past its cap.
- Duplicates a single-owner rule (brute-force, output discipline, env-reader).
- Adds `DO NOT` / `MUST NOT` / `NEVER` outside an Anti-Patterns block.

## Output

Concise change report:
- **Updated.** File + one-line summary of edit.
- **Skipped.** Notable findings intentionally not added, with brief reasoning.
- **No changes.** State explicitly when nothing warranted an update.

## Reference

- [STRUCTURE.md](reference/STRUCTURE.md) — directory layout requirements.
- [FRONTMATTER.md](reference/FRONTMATTER.md) — YAML rules.
- [CONTENT.md](reference/CONTENT.md) — writing guidelines.

## Anti-Patterns

- Creating CHANGELOG.md / SUMMARY.md / VERIFICATION.md auxiliary files.
- Meta-documentation about the creation process inside the skill itself.
- Verbose inline templates and examples (link to `reference/` instead).
- Re-introducing duplicate rule prose (brute-force, output-dir, env-reader).
- Files past their cap — split into `reference/` immediately.

讨论

基于 GitHub Discussions。登录 GitHub 即可参与讨论、点赞、订阅更新。