sglang-bisect-ci-regression

Bisect CI commit windows to diagnose regressions from failure logs.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the diagnosis of CI regressions by bisecting the commit window between passing and failing runs using scheduled main CI jobs.

Core Features & Use Cases

  • Extract failure signatures from CI logs
  • Identify the commit boundary between passing and failing runs
  • Correlate runner/hardware data to classify failures as code regressions, hardware-specific issues, or environment changes
  • Optionally reproduce failures on a remote GPU host when SSH target and container details are provided

Quick Start

Provide a failing test name or CI job URL, and optionally an SSH target and docker container to begin the bisection.

Frequently Asked Questions about sglang-bisect-ci-regression

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

FAQPage Schema
How do I find which commit caused a CI regression in GitHub Actions?

This skill bisects the commit window between passing and failing scheduled main-branch CI runs, extracting failure signatures from GitHub Actions logs to pinpoint the exact breaking commit.

Can I reproduce CI failures on a remote GPU host automatically?

Yes, you can reproduce CI failures on a remote GPU host by providing an SSH target and docker container details, allowing the skill to recreate the failing environment and validate the regression.

How does log analysis distinguish between code regressions and hardware-specific issues?

Log analysis distinguishes code regressions from hardware or environment issues by correlating runner and hardware data with extracted failure signatures, classifying the root cause based on the environment context.

Do I need a specific CI workflow setup to use automated bisecting?

You need scheduled main-branch CI workflows in GitHub Actions, as the bisection process analyzes the commit boundary between historical passing and failing runs triggered by these scheduled jobs.

What inputs are required to start diagnosing a CI regression?

To start diagnosing a CI regression, you need to provide a failing test name or CI job URL, with optional SSH target and docker container details for remote reproduction on a GPU host.