Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计提醒

check-ci检查 CI

Agent Skill

check-ci 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

729

周安装

31

GitHub Stars

37

下载量

255
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:check-ci(检查 CI)
来源仓库:https://github.com/majesticlabs-dev/majestic-marketplace
仓库路径:skills/check-ci
安装命令:
npx skills add https://github.com/majesticlabs-dev/majestic-marketplace --skill check-ci
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/majesticlabs-dev/majestic-marketplace --skill check-ci

简介

check-ci 轮询 GitHub Pull Request 的 CI 检查结果,直至完成或超时告警。

  • 支持自定义间隔与超时阈值,适用于等待构建通过再继续流程的场景。
  • 集成于自动化工作流,可结合 webhook 触发或定时巡检机制使用。
  • 需配置 GITHUB_TOKEN 并具有仓库只读权限以查询 PR 状态。
  • check-ci 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Check CI

Overview

This skill provides automated monitoring of GitHub Pull Request CI checks. It polls PR check status at regular intervals and reports when checks pass, fail, or timeout, enabling continuous integration workflow automation.

When to Use This Skill

Trigger this skill when the user requests:

  • Checking CI status for a PR
  • Monitoring CI checks until completion
  • Waiting for CI to pass before proceeding
  • Verifying build status
  • Polling GitHub PR checks with custom intervals or timeouts

Core Workflow

1. Execute CI Polling

Run the check_ci.sh script to monitor PR checks:

bash scripts/check_ci.sh [PR_NUMBER] [INTERVAL] [TIMEOUT]

Parameters:

  • PR_NUMBER (optional): PR number to monitor. Auto-detects from current branch if not provided
  • INTERVAL (optional, default: 15): Seconds between polling attempts
  • TIMEOUT (optional, default: 600): Maximum wait time in seconds

Examples:

# Auto-detect PR from current branch
bash scripts/check_ci.sh

# Monitor specific PR #123
bash scripts/check_ci.sh 123

# Custom interval (30s) and timeout (15 min)
bash scripts/check_ci.sh 123 30 900

2. Interpret Results

The script returns three possible outcomes:

✅ Success (exit code 0): All CI checks passed

  • Indicate that the PR is green and ready for next steps
  • Suggest proceeding with merge, review, or deployment

❌ Failure (exit code 1): One or more CI checks failed

  • Parse the output to identify specific failing checks
  • Analyze failure patterns (test failures, lint errors, build issues)
  • Suggest concrete fixes based on the failure type:

- Test failures: Review test output, fix failing tests - Lint errors: Run linter locally, apply fixes - Build failures: Check compilation errors, dependencies - Type errors: Review type definitions, fix type issues

⏰ Timeout (exit code 2): Polling exceeded timeout limit

  • Suggest running with longer timeout: check_ci.sh [PR] [interval] [longer_timeout]
  • Recommend checking GitHub directly for CI status
  • Consider whether CI is actually running or stuck

3. Provide Actionable Summary

Structure the response as follows:

## CI Polling Results for PR #[number]

### Final Status
[✅ All checks passed / ❌ Checks failed / ⏰ Polling timed out]

### Polling Summary
- Duration: [time]
- Final state: [description]
- Specific failures: [if applicable]

### Next Steps
[Specific actions based on results]

CI Failure Resolution

When checks fail, diagnose and fix by stack:

TECH_STACK = /majestic:config tech_stack generic

Failure Patterns by Stack

Ruby/Rails

FailureIndicatorsFix
TestFAILED, ErrorFix test or code
RubocopOffenses:bundle exec rubocop -a
TypesTypeErrorAdd nil checks
Bundlebundle install failedFix Gemfile

Verify: bundle exec rspec && bundle exec rubocop

Node.js

FailureIndicatorsFix
TestFAIL, expectedFix test or code
ESLintrule namesnpm run lint -- --fix
TypeScriptTS2xxxFix type errors
Depsnpm ERR!Fix package.json

Verify: npm test && npm run lint

Python

FailureIndicatorsFix
TestFAILED, AssertionErrorFix test or code
RuffFound x error(s)ruff check --fix
TypesmypyFix type hints
DepsModuleNotFoundErrorUpdate deps

Verify: pytest && ruff check.

Go

FailureIndicatorsFix
Test--- FAIL:Fix test or code
Lintlinter warningsgolangci-lint run --fix
Buildcannot find packagego mod tidy

Verify: go test./... && golangci-lint run

Resolution Workflow

  1. Fetch - Get failed workflow logs via gh run view <RUN_ID> --log-failed
  2. Parse - Identify failure type (test, lint, build, deps)
  3. Locate - Find relevant code files
  4. Fix - Make minimal, focused changes
  5. Verify - Run tests/linters locally

Resolution Principles

  • Fetch actual logs before assuming
  • Detect project type first
  • Make minimal changes
  • Verify locally before reporting
  • If unclear, state interpretation

Advanced Usage

Auto-Detection Requirements

For PR auto-detection to work:

  • Must be in a git repository
  • Current branch must have an associated PR
  • GitHub CLI (gh) must be authenticated

Error Handling

No PR found:

  • Verify current branch has an associated PR: gh pr view
  • Suggest creating a PR if none exists
  • Check if branch is pushed to remote

GitHub CLI issues:

  • Verify gh is installed and authenticated
  • Check repository access permissions
  • Ensure network connectivity

Integration with Other Workflows

This skill complements:

  • PR creation workflows (check CI after pushing)
  • Code review processes (verify CI before reviewing)
  • Deployment pipelines (ensure CI passes before deploy)
  • Automated testing workflows

Resources

scripts/check_ci.sh

Bash script that performs the actual CI polling. The script:

  • Auto-detects PR number from current branch or accepts as argument
  • Auto-detects repository owner/name from git remote
  • Polls gh pr checks at configurable intervals
  • Returns appropriate exit codes for success/failure/timeout
  • Provides real-time status updates during polling

The script can be executed directly without loading into context, making it token-efficient for repeated CI monitoring operations.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

34%
按下载量换算87

Claude

28.58%
按下载量换算73

Cursor

18.92%
按下载量换算48

Gemini CLI

9.43%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills