Gh Pr Checks Plan Fix
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions; use `gh` to inspect checks and logs, summarize failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example Buildkite) as out of scope and report only the details URL.
概览
The Gh Pr Checks Plan Fix skill, hosted in the mxyhi/ok-skills repository, enables AI agents to diagnose and resolve failing GitHub Pull Request checks. By leveraging the GitHub CLI, the skill inspects check statuses and retrieves execution logs to pinpoint the root cause of CI/CD failures within GitHub Actions. After summarizing the failure context, the agent drafts a remediation plan, though implementation only proceeds following explicit user approval. While the skill is optimized for native GitHub Actions, it identifies external providers like Buildkite by reporting their detail URLs without attempting direct debugging. This tool is compatible with agents such as Codex and Claude, utilizing the popularity of its source repository to streamline automated DevOps troubleshooting workflows.
使用场景
安装说明
# Review source first
open https://github.com/mxyhi/ok-skills/blob/main/gh-fix-ci/SKILL.mdCopy or clone the skill folder into your agent skills directory after reviewing its instructions and scripts.
安全提示
This skill requires GitHub CLI access and appropriate repository permissions to read logs and modify code. To maintain repository integrity, it enforces a mandatory manual approval step before any proposed fixes are implemented. Users must ensure that the AI agent environment is securely configured to manage GitHub authentication tokens.
相关 Skills
Protected Vercel Deployments
vercel-labs/agent-browser
使用 agent-browser 访问并测试受 Vercel Authentication、SSO 或 Deployment Protection 保护的 Vercel 部署。当预览或生产环境的 URL 重定向到 Vercel 登录页面、返回 401 或 403 保护错误，或者需要短期 Trusted Sources OIDC 身份验证（而非静态绕过密钥或公开例外）时，请使用此功能。
Agent Browser
vercel-labs/agent-browser
为 AI 智能体提供的浏览器自动化 CLI。当用户需要与网站交互时使用，包括页面导航、填写表单、点击按钮、截取屏幕截图、提取数据、测试 Web 应用或自动化任何浏览器任务。触发场景包括“打开网站”、“填写表单”、“点击按钮”、“截取屏幕截图”、“从页面抓取数据”、“测试此 Web 应用”、“登录网站”、“自动化浏览器操作”等请求，或任何需要程序化 Web 交互的任务。还可用于
Deep Agents Core
langchain-ai/langchain-skills
在构建 ANY Deep Agents 应用程序时的 INVOKE THIS SKILL。涵盖了 create_deep_agent()、harness 架构、SKILL.md 格式以及配置选项。
Dx Devops Test Pipeline Configure
forcedotcom/sf-skills
配置 DevOps Center 流水线测试基础设施：启用测试提供商以使其测试套件可用，重新同步已配置的提供商以拉取新套件，或在阶段上创建带有规则的质量门禁。在运行共享先决条件检查和显式确认门禁后，根据意图在三种模式之间进行路由。当用户想要设置、配置、启用、同步或刷新测试提供商，或者在 DevOps Center 流水线上设置/配置质量门禁或覆盖率阈值时，请使用此技能。
Agentforce Test
forcedotcom/sf-skills
编写、运行和分析 Agentforce 代理的结构化测试套件 —— 功能性 AND 安全性。TRIGGER 场景：用户编写或修改测试规范 YAML (AiEvaluationDefinition)；运行 sf agent test create、run、run-eval 或 results 命令；询问测试覆盖策略、指标选择或自定义评估；解释测试结果或诊断测试失败；询问批量测试、回归套件或 CI/CD 测试集成；请求安全测试、OWASP LLM Top 10、红队测试、渗透测试。
Provider Ephemeral Resources
hashicorp/agent-skills
使用 Plugin Framework 实现 Terraform provider ephemeral resources：Open/Renew/Close 生命周期、ephemeral schema 设计、通过 EphemeralResources 注册、过期凭据续期，以及 ephemeral values 如何流向 write-only 属性和 provider 配置。适用于添加 ephemeral resource、公开绝不能持久化在 state 或 plan 中的 secrets/tokens/certificates、在 ephemeral resource 和 data source 之间做出选择，或连接短期凭据。