git2linkedin

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Git2LinkedIn

Overview

概述

Generate a LinkedIn Experience draft from git history by directly reading commit messages and converting them into feature-focused, business-relevant, and confidentiality-safe bullets.

通过直接读取提交消息并将其转换为聚焦功能、与业务相关且符合保密要求的项目符号，从Git历史生成LinkedIn履历草稿。

Workflow

工作流程

Collect required inputs:

```
role
```
(required)
```
company
```
(required)

Collect optional inputs:

```
repo
```
(default
```
.
```
)
```
since
```
(
```
YYYY-MM-DD
```
)
```
until
```
(
```
YYYY-MM-DD
```
)
```
author
```
(default: current git user)
```
out
```
(optional markdown output path)
```
max-commits
```
(default
```
400
```
)

Run the extractor:

bash

undefined

收集必填输入项：

```
role
```
（必填）：职位
```
company
```
（必填）：公司

收集可选输入项：

```
repo
```
（默认值为
```
.
```
）：仓库路径
```
since
```
（格式为YYYY-MM-DD）：起始日期
```
until
```
（格式为YYYY-MM-DD）：结束日期
```
author
```
（默认值为当前Git用户）：提交作者
```
out
```
（可选）：Markdown输出路径
```
max-commits
```
（默认值为400）：最大提交数

运行提取脚本：

bash

undefined

All-time

全时段提取

python3 skills/git2linkedin/scripts/git_history_extract.py
--role "Senior Software Engineer"
--company "Acme"

Date range

按日期范围提取

python3 skills/git2linkedin/scripts/git_history_extract.py
--role "Senior Software Engineer"
--company "Acme"
--since 2025-01-01
--until 2025-12-31


4. Convert JSON output into final LinkedIn text:
- Write one short English summary sentence.
- Write 4-6 impact bullets.
- Focus on end-user value and business relevance.
- Avoid repository metrics, commit hashes, file paths, and internal identifiers.

5. Handle empty ranges explicitly:
- If commit list is empty, explain that no commits were found and suggest widening the date range or changing `author`.

python3 skills/git2linkedin/scripts/git_history_extract.py
--role "Senior Software Engineer"
--company "Acme"
--since 2025-01-01
--until 2025-12-31


4. 将JSON输出转换为最终LinkedIn文本：
- 撰写一句简短的英文总结句。
- 撰写4-6条聚焦成果的项目符号。
- 重点突出终端用户价值和业务相关性。
- 避免包含仓库指标、提交哈希、文件路径及内部标识符。

5. 显式处理空范围情况：
- 如果提交列表为空，需说明未找到任何提交，并建议扩大日期范围或修改`author`参数。

Output Contract

输出约定

Use these JSON fields:

```
summary
```
: baseline sentence for LinkedIn experience
```
feature_highlights
```
: feature-oriented bullet candidates
```
end_user_outcomes
```
: user-facing outcomes
```
business_relevance
```
: why this work matters for role/business goals
```
source_mode
```
: extraction mode metadata (
```
direct-commit-messages
```
)
```
guidance
```
: fallback suggestion when history is empty

使用以下JSON字段：

```
summary
```
：LinkedIn履历的基础总结句
```
feature_highlights
```
：聚焦功能的项目符号候选内容
```
end_user_outcomes
```
：面向终端用户的成果
```
business_relevance
```
：该工作对职位/业务目标的重要性
```
source_mode
```
：提取模式元数据（
```
direct-commit-messages
```
）
```
guidance
```
：当历史记录为空时的备选建议

Guardrails

约束规则

Keep output in English unless user asks otherwise.
Keep bullets concise and outcome-focused.
Keep sentence style mostly passive and neutral; avoid first-person phrasing (
```
I improved
```
,
```
I enhanced
```
, etc.).
Do not include repository statistics.
Do not expose commit hashes, file paths, issue IDs, URLs, or raw commit messages.
Do not invent confidential product details; keep language high-level and safe for public profiles.

除非用户另行要求，否则输出内容保持为英文。
项目符号需简洁且聚焦成果。
句子风格尽量采用被动、中立语气；避免使用第一人称表述（如
```
I improved
```
、
```
I enhanced
```
等）。
不得包含仓库统计数据。
不得暴露提交哈希、文件路径、问题ID、URL或原始提交消息。
不得编造机密产品细节；语言需保持高概括性，确保适合公开个人资料。