用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/ai-dynamo/aitune --skill aitune-release命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Use when a user asks to tune, optimize, accelerate, or deploy a PyTorch model or inference pipeline for GPU performance.
Use when measuring inference performance of a PyTorch model before and after AITune optimization — capturing baseline, compilation time, tuned throughput, and speedup.
Use when creating a git commit to ensure the message follows the Conventional Commits specification
基于 SOC 职业分类
正在显示 SKILL.md
| name | aitune-release |
| description | Use when creating a new release tag for the project |
| allowed-tools | Bash |
| license | Apache-2.0 |
Before tagging, verify:
Working tree is clean:
git status --short
If there are uncommitted changes, stop and ask the user to commit or stash them first.
Current branch is main and up to date:
git branch --show-current
git fetch origin && git status
Latest pipeline on main is passing — check in CI/CD pipelines before tagging. A tag pushed against a failing pipeline will trigger a broken deploy. Stop and ask the user to confirm the pipeline is green.
Read the latest tag from origin:
git fetch --tags origin
git tag --sort=-version:refname | head -5
Ask the user: major, minor, or patch?
Compute the next version:
v1.2.3v2.0.0, Minor → v1.3.0, Patch → v1.2.4Propose the tag (e.g. v1.3.0) and wait for confirmation.
On confirmation, create an annotated tag and push:
git tag -a v1.3.0 -m "Release v1.3.0"
git push origin v1.3.0
Pushing the tag triggers the GitLab CI tag pipeline automatically.