A simple local CLI for creating, installing, sharing, and using AI skills.
Project description
SKBro
Your local AI skill manager.
SKBro 是一个本地优先、零服务依赖的 AI Skill 管理工具,用来创建、安装、搜索、使用、更新和分享 Skill。
特点
- 支持完整 Skill 目录和单个 Markdown 文件;
- 支持本地目录、ZIP、HTTP URL 和 Git 仓库;
- 支持软链接和复制两种项目使用方式;
- 内置 Codex、Claude 和通用项目目录;
- 可以把 Skill 打包成 ZIP 分享;
- 本地 JSON 存储,无账号、数据库或后台服务;
- 核心只使用 Python 标准库。
安装
需要 Python 3.10 或更高版本。
PyPI 安装
正式发布到 PyPI 后,推荐使用:
pipx install skbro
或:
uv tool install skbro
从 GitHub 安装
pipx install "git+https://github.com/AuCf/skbro.git"
从本地源码安装
git clone https://github.com/AuCf/skbro.git
cd skbro
pipx install .
开发时也可以直接运行:
python -m skbro --help
验证安装:
skbro --version
skbro --help
更新或卸载:
pipx upgrade skbro
pipx uninstall skbro
最快上手
创建一个 Skill:
skbro create api-reviewer \
--description "Review API designs" \
--tags api,review
生成:
api-reviewer/
├── SKILL.md
└── skill.json
安装并用于当前项目:
skbro add ./api-reviewer
skbro use api-reviewer
默认位置:
.skills/api-reviewer/
用于 Codex 或 Claude:
skbro use api-reviewer --target codex
skbro use api-reviewer --target claude
安装别人分享的 Skill
# 本地目录或 Markdown
skbro add ./my-skill
skbro add ./my-skill.md --name my-skill
# ZIP
skbro add ./my-skill-1.0.0.zip
# HTTP URL
skbro add https://example.com/my-skill-1.0.0.zip
# Git
skbro add https://github.com/example/my-skill.git
skbro add https://github.com/example/my-skill.git --ref v1.0.0
日常管理
skbro list
skbro search API
skbro info api-reviewer
skbro status
skbro update api-reviewer
skbro unuse api-reviewer
skbro remove api-reviewer
skbro doctor
不指定名称时,skbro update 会更新所有有来源记录的 Skill。
软链接与复制模式
默认使用软链接。中央 Skill 更新后,项目立即生效:
skbro update api-reviewer
Windows 无法创建软链接,或希望项目保留完整副本时:
skbro use api-reviewer --copy
复制模式更新后需要同步:
skbro status
skbro sync api-reviewer
如果项目副本被人工修改,sync 和 unuse 会拒绝覆盖或删除;确认放弃修改时使用 --force。
分享 Skill
skbro pack api-reviewer
会生成:
api-reviewer-0.1.0.zip
对方安装:
skbro add ./api-reviewer-0.1.0.zip
也可以把 Skill 目录放进独立 Git 仓库分享。
Skill 格式
最小格式:
my-skill/
└── SKILL.md
推荐格式:
my-skill/
├── SKILL.md
├── skill.json
├── scripts/
├── references/
└── assets/
skill.json 示例:
{
"schema_version": 1,
"name": "api-reviewer",
"version": "1.0.0",
"description": "Review API designs",
"note": "用于检查接口一致性、安全性和可维护性",
"tags": ["api", "review"],
"author": "Your Name",
"license": "MIT",
"entry": "SKILL.md"
}
配置
skbro config
skbro config set link_mode copy
skbro config set default_target codex
skbro config set target.cursor .cursor/skills
默认数据目录:
~/.skbro/
├── config.json
├── registry.json
└── skills/
测试或多环境使用时,可设置 SKBRO_HOME。
旧版兼容
skm命令继续作为兼容别名;SKM_HOME环境变量继续可用;- 已存在
~/.skill_manager时会继续使用,不会丢失原数据; register/link/unlink旧命令继续可用;- 旧版单 Markdown registry 可以继续读取。
开发验证
python -B -m unittest discover -s tests -v
测试覆盖本地生命周期、旧版兼容、Markdown 转换、ZIP/HTTP/Git 安装、更新同步、修改保护、路径安全和健康检查。
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file skbro-0.3.0.tar.gz.
File metadata
- Download URL: skbro-0.3.0.tar.gz
- Upload date:
- Size: 21.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ae45f5ab6d521f94a23a9f89ca826817efb7ba82c9f88b92cdea3a9544aa3e99
|
|
| MD5 |
cf8db831829b714ad181301bfd920b69
|
|
| BLAKE2b-256 |
6a5b2d6f7f59022353c7db1e5f72433457a5a51def6909052901516c8c90c9c3
|
Provenance
The following attestation bundles were made for skbro-0.3.0.tar.gz:
Publisher:
publish.yml on AuCf/skbro
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
skbro-0.3.0.tar.gz -
Subject digest:
ae45f5ab6d521f94a23a9f89ca826817efb7ba82c9f88b92cdea3a9544aa3e99 - Sigstore transparency entry: 2142147825
- Sigstore integration time:
-
Permalink:
AuCf/skbro@5d1572c8c985fc258c50ca3765215265ad1da7ef -
Branch / Tag:
refs/heads/main - Owner: https://github.com/AuCf
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@5d1572c8c985fc258c50ca3765215265ad1da7ef -
Trigger Event:
workflow_dispatch
-
Statement type:
File details
Details for the file skbro-0.3.0-py3-none-any.whl.
File metadata
- Download URL: skbro-0.3.0-py3-none-any.whl
- Upload date:
- Size: 23.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.12
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8a4a97ada584e3c8683b36904e024458e53f24163a2423475c8a48300933e241
|
|
| MD5 |
ce22b109621440435bb6c917ef4eb83a
|
|
| BLAKE2b-256 |
eb9515eb64351c995b217940c15c45488ee1ca32600dd2b4ced67e7d7d28cabf
|
Provenance
The following attestation bundles were made for skbro-0.3.0-py3-none-any.whl:
Publisher:
publish.yml on AuCf/skbro
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
skbro-0.3.0-py3-none-any.whl -
Subject digest:
8a4a97ada584e3c8683b36904e024458e53f24163a2423475c8a48300933e241 - Sigstore transparency entry: 2142148160
- Sigstore integration time:
-
Permalink:
AuCf/skbro@5d1572c8c985fc258c50ca3765215265ad1da7ef -
Branch / Tag:
refs/heads/main - Owner: https://github.com/AuCf
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@5d1572c8c985fc258c50ca3765215265ad1da7ef -
Trigger Event:
workflow_dispatch
-
Statement type: