investigating-ci-failures — Claude Code skill: CI failures
The investigating-ci-failures Claude Code skill traces an individual CI failure to its likely cause. It separates PR-specific problems from default-branch breaks and flaky tests.
no license 188 installs ★ 101 updated 2026-11-16
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a claude-codeWhat investigating-ci-failures does
The skill examines a failed test or red CI run and classifies the failure as tied to a PR, the default branch, or a flaky test. For a default-branch incident, it traces the responsible commit, its author and PR, then checks whether a later commit fixed the problem.
It draws on engineering analytics tools, CI failure logs, and the engineering_analytics_ci_failures and engineering_analytics_ci_job_history views. The investigation compares branch patterns and job attempts; a CI rerun is not needed. Fingerprint analysis covers pytest only, so Jest, Playwright, and Cargo failures are examined through their raw log lines. Job history arrives through webhook sync and can lag.
What it helps with
- tell whether a PR introduced a CI failure
- trace a default-branch failure to a commit
- identify the author and PR behind a breaking commit
- distinguish a flaky test from a persistent failure
- check whether a later commit fixed the failure
Example prompts for the agent
- Investigate the CI failure in my PR and identify its cause.
- Find the commit that made this test fail on the default branch.
- Check whether this test fails intermittently or on every run.
How to install investigating-ci-failures
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a claude-code2Manually: put the files in place
~/.claude/skills/investigating-ci-failures/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/investigating-ci-failures/
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a codex2Manually: put the files in place
.agents/skills/investigating-ci-failures/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a cursor2Manually: put the files in place
.cursor/skills/investigating-ci-failures/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install PostHog/ai-plugin skills/investigating-ci-failures2Manually: put the files in place
.github/skills/investigating-ci-failures/SKILL.md1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a gemini-cli2Platform CLI
gemini skills install https://github.com/PostHog/ai-plugin --path skills/investigating-ci-failuresManually: put the files in place
.devin/skills/investigating-ci-failures/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about investigating-ci-failures
How do I install investigating-ci-failures in Claude Code?
Run npx skills add PostHog/ai-plugin --skill investigating-ci-failures -a claude-code.
What does investigating-ci-failures do?
It investigates one CI failure and determines whether it belongs to a PR, the default branch, or a flaky test. For a default-branch failure, it traces the commit and checks for a later fix.
What does investigating-ci-failures need to work?
It uses the engineering-analytics-broken-tests tool, CI failure log tools, and the engineering_analytics_ci_failures and engineering_analytics_ci_job_history views.