技能 diagnose-ci-failures

📦

diagnose-ci-failures

Name: diagnose-ci-failures
Author: warpdotdev

内容修订 r1 安全 ⚙️ 外部命令

Diagnose CI Failures Faster

CI failures can hide the first useful error among checks, logs, and repeated noise. This skill gathers relevant GitHub evidence and produces a prioritized plan for repair.

支持: Claude Codex Code(CC)

🥉 78 青铜

使用我的智能体安装

将此请求复制给你的智能体，其中包含权威 Skill 页面和清单。

智能体请求

Review the Skillstore skill "diagnose-ci-failures" from https://skillstore.io/skills/warpdotdev-diagnose-ci-failures.md and its manifest at https://skillstore.io/api/skills/warpdotdev-diagnose-ci-failures/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

你的智能体仍应展示其计划，并按安全策略请求必要确认。

Agent 可读资源

当 AI Agent、爬虫或脚本需要干净上下文，而不是读取完整页面时，请使用这些链接。

Markdown 详情 GET /skills/warpdotdev-diagnose-ci-failures.md 已签名 manifest GET /api/skills/warpdotdev-diagnose-ci-failures/manifest 已签名 lockfile GET /api/skills/warpdotdev-diagnose-ci-failures/lockfile

测试它

正在使用“diagnose-ci-failures”。 Diagnose the failures on pull request 42.

预期结果:

The Linux test job fails first because a fixture is missing. Formatting and downstream build checks may be secondary failures. Recommended order: restore the fixture, rerun tests, then address formatting.

正在使用“diagnose-ci-failures”。 Compare the failed macOS and Linux checks.

预期结果:

Both platforms report the same dependency resolution error.
The Windows job passed, so the issue may depend on runner configuration or lockfile handling.
Verify the dependency cache and lockfile before changing application code.

安全审计

安全

v1 • 2026/9/18 打开版本化报告

All 25 static detections are false positives: the reported backticks occur in Markdown examples, inline code formatting, or command references. The available evidence shows no executed command, injected input, data exfiltration, or prompt-injection content.

已扫描文件

114

分析行数

复核项

已忽略误报

最新一次完成的静态与语义审计未检测到已确认的安全发现。这并不能证明该 Skill 没有副作用。

审计者: claude

分享与引用此报告

分享版本化评估报告、中性徽章、嵌入卡片和引用。Skillstore 只报告证据，不判断该 Skill 是否安全。

打开版本化报告

复制报告链接

https://skillstore.io/skills/warpdotdev-diagnose-ci-failures/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 徽章

[![Skillstore security assessment](https://skillstore.io/badges/skills/warpdotdev-diagnose-ci-failures/security.svg)](https://skillstore.io/skills/warpdotdev-diagnose-ci-failures?utm_source=security_passport_badge)

HTML 徽章

<a href="https://skillstore.io/skills/warpdotdev-diagnose-ci-failures?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/warpdotdev-diagnose-ci-failures/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

嵌入卡片

<iframe src="https://skillstore.io/embed/skills/warpdotdev-diagnose-ci-failures.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

学术引用 (APA · BibTeX · CFF)

APA 引用

warpdotdev. (2026). diagnose-ci-failures security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/warpdotdev-diagnose-ci-failures/audits/1

BibTeX 引用

@techreport{warpdotdev-warpdotdev-diagnose-ci-failures-2026,
  author = {warpdotdev},
  title = {diagnose-ci-failures security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/warpdotdev-diagnose-ci-failures/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "diagnose-ci-failures security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "warpdotdev"
date-released: "2026-09-18"
url: "https://skillstore.io/skills/warpdotdev-diagnose-ci-failures/audits/1"
identifiers:
  - type: other
    value: "skillstore:warpdotdev-diagnose-ci-failures:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore 评分

评分依据证据置信度: 中

架构

可维护性

内容

社区

规范符合性

你能构建什么

Triage a pull request

Review failing checks on an active pull request and isolate the earliest actionable error.

Prepare a repair handoff

Convert noisy CI output into a prioritized plan that another engineer can implement.

Investigate cross-platform failures

Compare platform-specific checks and identify whether formatting, tests, builds, or tooling cause the failure.

试试这些提示

Check my pull request

Inspect the CI status for pull request <number> in <owner>/<repo>. Summarize failed checks and the first useful error for each.

Explain the failing logs

For pull request <number>, retrieve the relevant failed job logs, remove repeated noise, and explain the likely cause in plain language.

Build a fix plan

Analyze all failed checks for pull request <number>. Group related failures, rank them by likely root cause, and create a step-by-step repair plan with verification commands.

Compare platform failures

Investigate pull request <number> across macOS, Linux, and Windows checks. Distinguish shared root causes from platform-specific issues and cite the evidence for each conclusion.

最佳实践

Provide the repository and pull request number so the diagnosis targets the correct checks.
Fix the earliest root-cause failure before addressing downstream or repeated errors.
Review the proposed commands and changes before running them in a repository.

避免

Do not treat every failed check as an independent root cause.
Do not expose private logs or credentials in copied prompts or issue comments.
Do not rerun expensive workflows repeatedly without changing the suspected cause.

常见问题

What does this skill inspect?

It inspects pull request checks, failed workflow jobs, and accessible GitHub Actions logs.

Does it fix the code?

No. It explains the evidence and prepares a repair plan for an engineer to review and implement.

What access is required?

The GitHub CLI must be authenticated and authorized to read the repository and relevant workflow logs.

Can it diagnose local test failures?

It focuses on CI evidence. You can provide local output separately, but local execution is not its primary capability.

How does it handle many failures?

It groups related checks and prioritizes likely root causes before secondary failures.

Should I run suggested commands immediately?

No. Review each command, its scope, and its repository impact before execution.

开发者详情

作者

warpdotdev

许可证

MIT

Skillstore 修订

版本提示

作者未声明版本。

仓库

https://github.com/warpdotdev/common-skills/tree/69b4753651ab7fab518c82be087b9f1d5b966631/.agents/skills/diagnose-ci-failures

引用

6e663c781098ffdba3ad66055d217e0dd327bd20

维护新鲜度

2026/9/18

使用量

0 次下载 · 0 次浏览

文件结构

📄 SKILL.md

warpdotdev 的更多技能

查看全部

📦

council

Compare Decisions with an AI Council

Complex decisions can suffer from narrow analysis and hidden assumptions. This skill coordinates diverse agents, compares their evidence, and produces one actionable recommendation.

效率查看

📦

write-product-spec

Write Clear Product Specifications

Turn ambiguous feature requests into structured product specifications that align design, engineering, and review. This skill defines user-visible behavior, scope, references, open questions, and handoff details in a consistent PRODUCT.md format.

效率查看

📦

spec-driven-implementation

Plan Features With Living Specs

Substantial feature work can become ambiguous when product intent, technical decisions, and implementation drift apart. This skill guides agents through practical product and technical specifications that stay aligned with shipped work.

编码与开发查看

📦

migrate-to-factory

Migrate Workflows to Factory

Moving an existing agent workflow into Factory can create duplicate roles, broken paths, or conflicting handoffs. This skill guides a focused migration into an existing Factory repository and validates the resulting configuration.

DevOps 与基础设施查看

📦

respond-to-pr-comments-in-blocklist

Resolve Pull Request Feedback Safely

Pull request feedback can be difficult to track, implement, and answer consistently. This skill guides each decision, validates accepted changes, and previews GitHub responses before posting.

编码与开发查看

📦

research

Delegate Broad Codebase Research

Large investigations can flood the primary agent context with irrelevant files and logs. This skill delegates read-only research and returns concise conclusions with cited evidence.

研究与学习查看

查看全部

📦