github-triage

Automate GitHub issue triage by fetching open items, classifying them into actionable categories, and spawning independent background subagents to analyze and respond in parallel. This approach enables scalable parallel processing for high-volume repositories.

The best way to triage pull requests is to fetch, categorize, and delegate them to background agents for parallel processing. This method coordinates results via task updates to deliver a live triage dashboard and a final summary for consistent routing.

Automated GitHub triage requires a YAML frontmatter with name and description for discovery, and a Markdown body with operational instructions. Optional directories like scripts, references, and assets can be included to support deterministic tasks and documentation.

Yes, you can use background tasks to process GitHub issues in parallel. The system spawns independent subagents per item to perform analysis, comments, or flag actions, coordinating the results to deliver a live triage dashboard and final summary.

Background subagent delegation works by fetching open issues and PRs, classifying them into categories, and spawning independent background tasks per item. These tasks perform analysis and route actions, using TaskCreate and TaskUpdate to coordinate a live triage dashboard.

A limitation of using parallel agents for GitHub triage is that it requires a Markdown body with operational instructions and optional directories for deterministic task support. Without proper structural setup, the background subagents cannot consistently analyze and route items.