skillreg
本地 AI Agent Skill 管理工具
skillreg 用来管理本地 AI Agent Skills。你可以把任意项目里写好的
SKILL.md 注册到统一 workspace，再同步到 Claude、Codex、CodeBuddy 或
~/.agents/skills 等 agent skill 目录。
能力
- CLI：创建/切换 workspace，注册/转换 skill，管理 NPM 来源、targets/projects，执行 sync，查看 diff，启动 dashboard。
- Skill：内置
skillreg-skill，让 agent 在任意项目里识别本地SKILL.md并注册到当前 workspace。 - Dashboard：提供图形入口，可从本地、ZIP、Git 或 NPM 包导入 skill，管理 NPM 来源、同步目标、项目组、仓库状态、依赖图与 Git 记录。
安装与使用
uv tool install skillreg
skillreg workspace create ~/my-skills
skillreg register /path/to/my-skill
skillreg source npm import @scope/skill-pack --registry https://registry.npmjs.org/ --mode skill
skillreg target add ~/.codex/skills
skillreg sync execute --target ~/.codex/skills
skillreg dashboard open
Dashboard 默认打开：
http://127.0.0.1:28787
常用命令：
skillreg config
skillreg workspace create <path>
skillreg workspace current
skillreg workspace switch <path>
skillreg register <path> [--force] [--name name]
skillreg list
skillreg convert <name>
skillreg source npm preview <package> [--registry url] [--version-spec spec]
skillreg source npm import <package> [--mode skill|repo] [--skill name]
skillreg source list
skillreg source check <source-id>
skillreg source update-preview <source-id>
skillreg source update <source-id> [--dry-run] [--force]
skillreg target list
skillreg target add <path>
skillreg sync status
skillreg sync execute --target <path> [--skill name]
skillreg project create --name <name> --target <path>
skillreg sync execute --project <name>
skillreg diff <skill> --target <path>
skillreg submodule list
skillreg dashboard start
skillreg dashboard status
skillreg dashboard stop
skillreg dashboard open --no-browser
NPM 来源以包为管理单元。一份包可选择导入多个 Skill 到 skills/，也可在
repos/ 下创建独立集合仓。skillreg 直接下载并校验 registry 制品，不执行
npm install 或生命周期脚本；来源、版本、完整性和目录映射保存在
.skillreg/sources.json，后续可检查版本、预览逐文件差异并受控更新。
开发
git clone https://github.com/fcraft/skillreg.git
cd skillreg
uv sync --extra dev
scripts/install-git-hooks.sh
uv run pytest -q
uv run --with ruff ruff check src/ tests/ scripts/
版本号由 pyproject.toml、src/skillreg/__init__.py 和内置
skillreg-skill 共同校验。安装 git hook 后，在 main 分支提交时会自动
bump x.y.z 版本：feat: 提交使 y + 1，其他提交使 z + 1，并把
x 固定为 1。
本地发版：
scripts/release.sh
脚本会读取当前版本，创建并推送 v<version> tag，随后由 GitHub Actions
发布到 GitHub Release 和 PyPI。
Contributors
Trend
License
1 maintainer
Metadata
1 maintainer
Release files for skillreg 1.6.5
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| skillreg-1.6.5.tar.gz | 332.3 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| skillreg-1.6.5-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 553.6 kB
Release files / skillreg-1.6.5.tar.gz
| Download URL | skillreg-1.6.5.tar.gz |
|---|---|
| Size | 332.3 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
28d48f50040785110d4ec71691c2f6a5d49fe21028fc875026bb1284c8e2a78b
|
|
BLAKE2b-256 checksum How to use checksums |
8e845d6e24220ee0b12936ee2849d73d531040324ecbbc65bf1bd7d40f4734f7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 22, 2026.
Transparency logRelease files / skillreg-1.6.5-py3-none-any.whl
| Download URL | skillreg-1.6.5-py3-none-any.whl |
|---|---|
| Size | 221.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
5de49bb011815f0e70a42a5ea24bdb6dc71e1e1a806b8abcbd7d14fcd159ced6
|
|
BLAKE2b-256 checksum How to use checksums |
ed1869278c586ef0304ec8d12ee852905f60239b666bf9ea542eb2f04223f073
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/6.1.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Jul 22, 2026.
Transparency log
