Menu
Diagnose and fix CI failures on PRs — automated test failure analysis and code fixes.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Use when you have an existing component, flow, or interface and need an evidence-backed accessibility design review after basic checks pass. Best for focus management, ARIA pattern quality, semantics, and state communication gaps automated tools miss.
Use when you know what component, flow, or interface you need but not yet the right accessibility approach. Best for turning requirements into an accessible implementation plan before code hardens bad interaction and state patterns.
Review content for Google AI Overview eligibility — RAG retrievability, E-E-A-T signal completeness, fan-out coverage, semantic HTML for agent parsability, schema markup quality.
Plan AI Overview readiness improvements — RAG retrievability, E-E-A-T signals, query fan-out coverage, semantic HTML, schema markup, agentic channel setup.
Assess organizational AI readiness — capabilities, gaps, adoption roadmap for strategic planning.
Alex Urevick-Ackelsberg's personal writing voice and style. Use this skill whenever writing AS Alex or ghostwriting content that should sound like him — emails to clients, community posts, conference session descriptions, proposals, LinkedIn posts, listserv replies, internal strategy docs, or any communication where Alex is the named author. Also use when Alex asks you to 'write this up,' 'draft a response,' 'help me write,' or when the output needs his voice rather than a generic professional tone. This is Alex's PERSONAL voice — for Zivtech brand/marketing content, use zivtech-writing-style instead (though both can apply when Alex is writing on behalf of Zivtech).
| name | ci-fix-executor |
| description | Diagnose and fix CI failures on PRs — automated test failure analysis and code fixes. |
| version | 0.1.0 |
Diagnose and fix CI failures so PRs can pass without manual debugging.
When CI has failed on a PR and I need the failure diagnosed and fixed, I want automated diagnosis, targeted fix generation, and local validation, so the PR can pass CI without me manually parsing log output.
qa-critic identified a flaky test pattern and I need to fix the underlying cause.test-plannerqa-criticperf-critic| User's Situation | What Happens | What They Leave With |
|---|---|---|
| Lint failure on PR | Executor parses lint output, identifies violations, applies auto-fix or generates targeted changes | A minimal diff fixing lint violations |
| Test failure on PR | Executor identifies the failing assertion, determines if test or implementation is wrong, generates targeted fix | A fix for the root cause (implementation fix or test correction with justification) |
| Build failure | Executor parses build output, identifies missing imports/config/deps, generates fix | A buildable state with the minimum change needed |
| Deploy failure | Executor identifies env/config/permission issues from deploy logs | Config fix or clear escalation with diagnosis |
| Dependency conflict | Executor resolves version constraints, updates lockfile | Resolved dependency tree |
test-plannertest-planner (optional upstream): provides testing strategy contextqa-critic (downstream): reviews the fix for correctness and test health impact