issue-bulk-process

Orchestrates parallel triage and serial reproduction of multiple GitHub issues with consolidated reporting.

5.6k|645|Updated Feb 22, 2016
One-click install
npx skills add https://github.com/mono/SkiaSharp --skill issue-bulk-process
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: issue-bulk-process
Source: https://github.com/mono/SkiaSharp/tree/main/.agents/skills/issue-bulk-process
Command: npx skills add https://github.com/mono/SkiaSharp --skill issue-bulk-process

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Processing GitHub issues one at a time is slow and repetitive when a repository accumulates a large backlog. This Skill automates bulk triage and reproduction of multiple SkiaSharp issues in a single command, delegating work to sub-agents and consolidating the results.

Core Features & Use Cases

  • Flexible Issue Selection: Accepts explicit issue numbers, "newest N issues" requests, or search/label-based queries resolved via the GitHub CLI.
  • Parallel Triage, Serial Reproduction: Runs read-only triage agents in parallel batches of up to 12, then runs reproduction agents sequentially to avoid build and simulator conflicts.
  • Skip and Force Modes: Detects previously processed issues from cached result files and skips them unless the user requests a forced re-run.
  • Consolidated Summary: Produces a single results table with type, severity, triage action, and reproduction conclusion per issue, plus aggregate stats.
  • Use Case: A maintainer returns from vacation to 30 new issues and runs one command to triage all of them, reproduce the bugs, and get a prioritized summary table.

Quick Start

Ask the assistant to bulk process issues 3400, 3428, and 3429, or to triage and reproduce the newest 10 open issues in the repository.

Frequently Asked Questions about issue-bulk-process

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

FAQPage Schema
How do I triage multiple GitHub issues at once?

Provide explicit issue numbers, a count like "newest 10 issues", or a search query, and the Skill resolves the list with the gh CLI. It then launches parallel triage agents in batches of up to 12 and reports a consolidated summary.

How do I reproduce bugs from a list of GitHub issues?

After triage, the Skill launches one reproduction agent per issue sequentially, since reproduction may build projects or open simulators. Each agent reports a conclusion such as reproduced, not-reproduced, needs-platform, or inconclusive.

Can I process only the newest open GitHub issues?

Yes. Phrases like "newest N issues" or "last N issues" trigger a gh issue list query that fetches the N most recent open issues. You can also filter by search terms or labels such as area/SkiaSharp.Views.

Does bulk issue processing skip issues that were already triaged?

Yes. The Skill checks for existing triage and reproduction JSON files in the .data-cache directory and skips those issues by default. Saying "force", "re-triage", or "redo" reprocesses everything.

What happens if a triage or reproduction agent times out?

Each triage agent gets up to 600 seconds total across two read attempts. Agents that time out are recorded with a timeout marker in the summary, and processing continues with the remaining issues.

Why does reproduction run serially instead of in parallel?

Reproduction may build projects, create files, or open simulators, which can conflict when run concurrently. Triage is read-only and safe to parallelize, but reproduction agents run one at a time to avoid interference.