D
dblock/already-tested
Skip duplicate CI tests using trusted, content-matched claims in pull request descriptions
在 GitHub 查看 ↗链接已于 2026-11-12 核查 · HTTP 200 · 打开的是厂商自己的官网
- ★ 数取值
- 0 ★ · dblock/already-tested · 取自该仓库的 GitHub 页面 · (2026-11-13 20:00:50 核对)
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型:开源项目
- 收录与活跃
- 收录于 2026-11-12 · 最近见到 2026-11-14 · 共见到 3 次 · 被 2 个来源列出
- 评分构成
- 综合分 203 · 2 个来源 · repo
常见问题(答案都来自我们的记录)
它适合谁 / 用来做什么?
我们把它归在这些标签下: 开源项目。标签来自条目自身,不是我们强加的分类。
谁维护它、还在更新吗?
仓库
dblock/already-tested;我们收录于 2026-11-12;最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。# already-tested
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
Skip duplicate CI tests when a trusted local run matches the committed source and workflow. Claims are not proof that tests ran.
Use your project's existing checks, in any language.
## Install
Git 2.29+ is required. Until standalone distribution is available, install the unreleased CLI from source using Node.js 24.x or 26+ and npm:
sh
git clone https://github.com/dblock/already-tested.git ../already-tested
npm install --global --ignore-scripts ../already-tested
This installs already-tested on your PATH, outside your project's dependencies. Node/npm are current installation details; the command stays the same with a future binary.
## Configure CI
Add the gate before setup and tests in .github/workflows/test.yml. Here ./script/ci stands for your project's existing CI command:
yaml
name: Test
on:
push:
pull_request:
types: [opened, synchronize, reopened, edited]
permissions:
contents: read
jobs:
test:
runs-on: ubuntu-latest
steps:
uses: actions/checkout@v7.0.1
with:
fetch-depth: 2
persist-credentials: false
uses: dblock/already-tested@master
id: tested
with:
check: test-linux
if: steps.tested.outputs.skip != 'true'
run: ./script/ci
Apply the same condition to your language's setup/install steps. The Action's runtime is provided by GitHub; it needs no setup-node step. Use a distinct check per job or matrix environment and pin a reviewed Action commit for production.
Missing, stale, or invalid claims run CI normally; pushes always run CI. A match skips the conditioned steps, not the workflow or its required check.
## Claim a Local Run
For the example above, use the matching Linux environment and replace the environment placeholder with the toolchain versions and OS you actually validated:
1. Run ./script/ci.
2. Commit the tested source without further edits.
3. Generate the claim:
sh
already-tested claim --workflow .github/workflows/test.yml --check test-linux \
--command './script/ci' --environment ' '
Paste the entire output into the PR description. It shows expandable validation details and hides the machine-readable claim. Regenerate after source or workflow changes. The CLI records your command and environment; it does not run or verify them.
For agents, add to your project's AGENTS.md:
markdown
Run CI checks locally in their matching environments, commit the tested source, then generate claims with already-tested claim --workflow --check --command ' ' --environment ' '. Paste the complete output into the PR description. Regenerate after changes; never claim checks or environments you did not validate.
Use your project's existing checks, in any language.
## Install
Git 2.29+ is required. Until standalone distribution is available, install the unreleased CLI from source using Node.js 24.x or 26+ and npm:
sh
git clone https://github.com/dblock/already-tested.git ../already-tested
npm install --global --ignore-scripts ../already-tested
This installs already-tested on your PATH, outside your project's dependencies. Node/npm are current installation details; the command stays the same with a future binary.
## Configure CI
Add the gate before setup and tests in .github/workflows/test.yml. Here ./script/ci stands for your project's existing CI command:
yaml
name: Test
on:
push:
pull_request:
types: [opened, synchronize, reopened, edited]
permissions:
contents: read
jobs:
test:
runs-on: ubuntu-latest
steps:
uses: actions/checkout@v7.0.1
with:
fetch-depth: 2
persist-credentials: false
uses: dblock/already-tested@master
id: tested
with:
check: test-linux
if: steps.tested.outputs.skip != 'true'
run: ./script/ci
Apply the same condition to your language's setup/install steps. The Action's runtime is provided by GitHub; it needs no setup-node step. Use a distinct check per job or matrix environment and pin a reviewed Action commit for production.
Missing, stale, or invalid claims run CI normally; pushes always run CI. A match skips the conditioned steps, not the workflow or its required check.
## Claim a Local Run
For the example above, use the matching Linux environment and replace the environment placeholder with the toolchain versions and OS you actually validated:
1. Run ./script/ci.
2. Commit the tested source without further edits.
3. Generate the claim:
sh
already-tested claim --workflow .github/workflows/test.yml --check test-linux \
--command './script/ci' --environment ' '
Paste the entire output into the PR description. It shows expandable validation details and hides the machine-readable claim. Regenerate after source or workflow changes. The CLI records your command and environment; it does not run or verify them.
For agents, add to your project's AGENTS.md:
markdown
Run CI checks locally in their matching environments, commit the tested source, then generate claims with already-tested claim --workflow --check --command ' ' --environment ' '. Paste the complete output into the PR description. Regenerate after changes; never claim checks or environments you did not validate.
收录徽章(可选)
想把收录标记放进 README 就复制这行,可选,不影响收录。
[](https://anclum.com/zh/s/dblock-already-tested/)
- 来源
- github · https://github.com/dblock/already-tested
- 按领域浏览
- 开源项目
- 收录来源
- 2 个独立来源
- 收录于
- 今天
同类推荐
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。