Loading...
Loading...
Compare original and translation side by side
sales-skills/salessales-skills/salesgh pr creategh issue createsales-skills/salessales-skills/salesgh pr creategh issue create$ARGUMENTSskills/sales-do/SKILL.md~/.claude/skills/$ARGUMENTSskills/sales-do/SKILL.md~/.claude/skills/Would you like to:
- Build the skill — I'll help you create it with proper structure and prepare a PR
- Request the skill — I'll file a GitHub issue so the maintainers know it's needed
您希望:
- 构建该Skill —— 我将帮助您按照规范结构创建Skill并准备PR
- 请求该Skill —— 我将在GitHub提交Issue,让维护者了解该需求
/skill-creator/skill-creatorRepo conventions for this skill:
- Naming:
for sales skills, descriptive names for marketing/GTM skills (e.g.,sales-<problem>,cold-email)launch-strategy- Descriptions should use phrases salespeople and marketers actually say — "write a cold email", "prep for a discovery call", "handle this objection"
- Description must end with negative triggers:
Do NOT use for X (use /alternative)- SKILL.md is the only required file — keep it focused and actionable
- Skills should ask clarifying questions before acting (audience, stage, constraints)
- Skills route through
— the description field determines when the router matches/sales-doSkill structure:skills/<skill-name>/ ├── SKILL.md # Main instructions (required) ├── scripts/ # Deterministic operations (data fetching, validation, formatting) ├── references/ # Large reference material (>500 words — API docs, data models) ├── assets/ # Templates, examples, configuration files └── evals/ └── evals.json # Test cases (optional, generated by skill-creator or manually)SKILL.md body pattern (follow what other skills in this repo do):
- Step to gather context (ask 2-4 questions about the user's specific situation)
- Implementation steps with actionable output
- Templates or frameworks relevant to the problem domain
- Gotchas section with 3-5 common Claude failure points for this domain
- Output formatting guidance
- Next steps pointing to related skills
Key principles:
- Don't state the obvious: Focus on info Claude wouldn't know — internal conventions, domain gotchas, non-obvious patterns
- Avoid railroading: Use "typically" instead of "always". Give Claude flexibility to adapt to the situation.
- Scripts: If the skill involves deterministic operations (data fetching, formatting, validation), include scripts in
scripts/- Progressive disclosure: Move reference material >500 words to
directoryreferences/
/skill-creator/skill-creator本仓库的Skill规范:
- 命名:销售类Skill使用
格式,营销/GTM类Skill使用描述性名称(例如sales-<问题>、cold-email)launch-strategy- 描述应使用销售和营销人员实际会说的表述——“撰写开发信”、“准备发现性通话”、“处理异议”
- 描述必须包含负面触发条件:
Do NOT use for X (use /alternative)- 仅需必备文件SKILL.md——内容需聚焦且可执行
- Skill应先提出澄清问题再执行操作(受众、阶段、约束条件)
- Skill通过
路由——描述字段决定路由是否匹配/sales-doSkill结构:skills/<skill-name>/ ├── SKILL.md # 主说明文档(必填) ├── scripts/ # 确定性操作(数据获取、验证、格式化) ├── references/ # 大型参考资料(超过500字——API文档、数据模型) ├── assets/ # 模板、示例、配置文件 └── evals/ └── evals.json # 测试用例(可选,由skill-creator生成或手动创建)SKILL.md正文模板(参考仓库中其他Skill的格式):
- 收集上下文的步骤(询问2-4个关于用户具体场景的问题)
- 可执行的实现步骤
- 与问题领域相关的模板或框架
- 注意事项部分:列出3-5个该领域中Claude常见的失误点
- 输出格式指导
- 指向相关Skill的下一步建议
核心原则:
- 不陈述显而易见的内容:聚焦Claude不知道的信息——内部规范、领域注意事项、非通用模式
- 避免强制流程:使用“通常”而非“必须”,给Claude适应场景的灵活性
- 脚本:如果Skill涉及确定性操作(数据获取、格式化、验证),将其放在
目录中scripts/- 渐进式披露:超过500字的参考资料移至
目录references/
---
name: <skill-name>
description: "<What problem it solves>. Use when <trigger phrases the user would say>. Do NOT use for <X> (use /alternative)."
argument-hint: "[brief hint about expected arguments]"
license: MIT
metadata:
author: sales-skills
version: 1.0.0
---skills//sales-doundefined---
name: <skill-name>
description: "<解决的问题>。当<用户会说的触发短语>时使用。Do NOT use for <X> (use /alternative)."
argument-hint: "[对预期参数的简要提示]"
license: MIT
metadata:
author: sales-skills
version: 1.0.0
---skills//sales-doundefined
**Body** — Should follow the question-first pattern: gather context about the user's situation before producing output. Include templates, frameworks, or examples that make the output immediately useful.
**正文**——应遵循先提问的模式:在生成输出前收集用户场景的上下文。包含能让输出立即生效的模板、框架或示例。evals/evals.json{
"skill_name": "<skill-name>",
"evals": [
{
"id": 0,
"prompt": "realistic user prompt a salesperson or marketer would say",
"expected_output": "description of what a successful response looks like",
"assertions": [
{"name": "assertion_name", "description": "specific thing to check in the output"}
]
}
]
}/skill-creatorevals/evals.json{
"skill_name": "<skill-name>",
"evals": [
{
"id": 0,
"prompt": "销售或营销人员会说的真实请求",
"expected_output": "合格输出的描述",
"assertions": [
{"name": "断言名称", "description": "输出中需要检查的具体内容"}
]
}
]
}/skill-creatorskills/sales-do/SKILL.mdREADME.mdgit checkout -b add-<skill-name>git add skills/<skill-name>/ evals/ skills/sales-do/SKILL.md README.md && git commit -m "Add <skill-name> skill"git push -u origin add-<skill-name>gh pr create \
--repo sales-skills/sales \
--title "Add <skill-name> skill" \
--body "$(cat <<'EOF'skills/sales-do/SKILL.mdREADME.mdgit checkout -b add-<skill-name>git add skills/<skill-name>/ evals/ skills/sales-do/SKILL.md README.md && git commit -m "Add <skill-name> skill"git push -u origin add-<skill-name>gh pr create \
--repo sales-skills/sales \
--title "Add <skill-name> skill" \
--body "$(cat <<'EOF'/<skill-name> <example prompt>/<skill-name> <示例请求>skills/<skill-name>/SKILL.mdskills/sales-do/SKILL.mdREADME.md
Return the PR URL to the user when done.skills/<skill-name>/SKILL.mdskills/sales-do/SKILL.mdREADME.md
完成后将PR链接返回给用户。gh issue create \
--repo sales-skills/sales \
--title "Skill request: <skill-name>" \
--body "$(cat <<'EOF'gh issue create \
--repo sales-skills/sales \
--title "Skill request: <skill-name>" \
--body "$(cat <<'EOF'
Return the issue URL to the user when done.
完成后将Issue链接返回给用户。sales-<problem>license: MITmetadata: { author, version }## Gotchasreferences/scripts/sales-do/SKILL.mdevals/evals.jsonsales-<问题>license: MITmetadata: { author, version }## Gotchasreferences/scripts/sales-do/SKILL.mdevals/evals.json/sales-donpx skills add sales-skills/sales --skills sales-do/sales-donpx skills add sales-skills/sales --skills sales-do