ci-workflow-guide

Diagnose and fix CI workflow orchestration issues across SGLang stages.

1|Updated May 8, 2026
One-click install
npx skills add https://github.com/dyyoungg/sglang-dev --skill ci-workflow-guide-dyyoungg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ci-workflow-guide
Source: https://github.com/dyyoungg/sglang-dev/tree/main/.claude/skills/ci-workflow-guide
Command: npx skills add https://github.com/dyyoungg/sglang-dev --skill ci-workflow-guide-dyyoungg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Streamlines SGLang CI workflow orchestration, enabling reliable stage sequencing, gating, and fast-fail debugging to reduce pipeline toil.

Core Features & Use Cases

  • Stage ordering and gating: defines the sequence of CI stages, how tests are dispatched, and how failures fast-fail across stages.
  • Cross-stage health checks: provides a fast-fail mechanism that surfaces root-cause failures while skipping cascading issues.
  • Architecture and references: covers core workflows, actions, and utilities involved in PR and scheduled CI runs.
  • Debugging CI failures: explains how to diagnose and resolve pipeline issues, including test partitioning and gating logic.
  • Use cases: modify CI workflows, add stages, debug pipeline issues, and understand test dispatch and gating across stages.

Quick Start

Follow this guide to implement a new CI stage by duplicating an existing stage, wiring it into check-changes and wait-for-stage, and validating with a test run.

Frequently Asked Questions about ci-workflow-guide

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug CI workflow failures in GitHub Actions?

Debug CI workflow failures by analyzing test dispatch, partitioning, and gating logic across stages. The guide provides debugging utilities and fast-fail mechanisms to surface root-cause failures while skipping cascading issues.

How does fast-fail logic work in CI pipeline orchestration?

Fast-fail logic in CI orchestration surfaces root-cause failures immediately while skipping downstream cascading issues. Cross-stage health checks identify the originating failure point to reduce pipeline toil.

What is the best way to add a new stage to an existing CI workflow?

Add a new CI stage by duplicating an existing stage, wiring it into check-changes and wait-for-stage logic, and validating the pipeline with a test run to ensure correct sequencing and gating.

How are tests dispatched and partitioned across CI stages?

Tests are dispatched and partitioned across CI stages according to defined stage ordering and gating policies. The orchestration logic specifies execution modes for both PR and scheduled runs.

Can I use this CI workflow guide for scheduled runs or only PR pipelines?

This guide applies to both PR and scheduled CI runs. It specifies stage ordering, gating policies, partitioning, and execution modes across different pipeline triggers for SGLang projects.

Why are CI stages failing out of order during pipeline execution?

CI stages failing out of order often indicates issues with wait-for-stage wiring or gating logic. Diagnose the orchestration by checking stage sequencing and cross-stage health check configurations.