turning-engineering-analytics-into-insights — Claude Code skill
turning-engineering-analytics-into-insights — a skill for Claude Code and Codex: converts PR and CI metrics into saved PostHog insights, dashboards, and automated subscriptions. Triggers when visualizing engineering team performance.
no license 189 installs ★ 101 updated 2026-11-16
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a claude-codeWhat turning-engineering-analytics-into-insights does
This skill guides the assistant in transforming repository and pipeline records into persistent PostHog charts via HogQL. Because MCP tool responses cannot be directly converted into saved insights, the workflow locates relevant warehouse tables using engineering-analytics-sources, tests the SQL via execute-sql, and registers the result through insight-create.
Queries respect PostHog data conventions. Timestamps arrive as strings requiring parseDateTimeBestEffort, nullable JSON fields require ifNull, and PR associations rely on run metadata instead of head SHA joins. Pre-built views like engineering_analytics_job_costs cover runner expenses, including runs processed by Depot CI.
Standard threshold alerts in PostHog only function with trend insights rather than SQL charts, so automated checks require scheduled deliveries or AI subscriptions. Specialized metrics such as flaky test rankings belong exclusively to dedicated MCP tools and must not be reconstructed manually in HogQL.
What it helps with
- Build a dashboard tracking PR review times and merge durations
- Schedule weekly email updates for team pull request throughput
- Construct validated HogQL queries against GitHub warehouse sources
- Save custom charts tracking pytest failure signatures over recent runs
Example prompts for the agent
- Put CI health and merge times on a new dashboard
- Save this pull request throughput metric as an insight
- Set up a weekly subscription for CI job costs
How to install turning-engineering-analytics-into-insights
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a claude-code2Manually: put the files in place
~/.claude/skills/turning-engineering-analytics-into-insights/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/turning-engineering-analytics-into-insights/
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a codex2Manually: put the files in place
.agents/skills/turning-engineering-analytics-into-insights/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a cursor2Manually: put the files in place
.cursor/skills/turning-engineering-analytics-into-insights/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install PostHog/ai-plugin skills/turning-engineering-analytics-into-insights2Manually: put the files in place
.github/skills/turning-engineering-analytics-into-insights/SKILL.md1Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a gemini-cli2Platform CLI
gemini skills install https://github.com/PostHog/ai-plugin --path skills/turning-engineering-analytics-into-insightsManually: put the files in place
.devin/skills/turning-engineering-analytics-into-insights/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -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 turning-engineering-analytics-into-insights
How to install turning-engineering-analytics-into-insights in Claude Code?
Run npx skills add PostHog/ai-plugin --skill turning-engineering-analytics-into-insights -a claude-code in your terminal.
What does turning-engineering-analytics-into-insights do?
It directs the agent to query GitHub warehouse tables with HogQL, format SQL insights properly, and organize them into PostHog dashboards or subscriptions.
What is required for turning-engineering-analytics-into-insights to work?
It requires synced GitHub workflow runs and jobs within PostHog warehouse storage alongside tools like engineering-analytics-sources and insight-create.