What problem does it solve?
Manually triaging incoming issues requires tracing code history, understanding architecture, and diagnosing root causes, which is slow and inconsistent across maintainers. This Skill automates the full investigation of a GitHub or Linear issue and produces a standardized triage verdict with severity, effort, and impact classifications.
Core Features & Use Cases
- End-to-end issue investigation: Parses the issue reference, reads the full thread, searches related issues and PRs, and traces the symptom through the codebase using git history and blame.
- Structured diagnosis and handoff: Produces a marker-prefixed comment with type, route, severity, confidence, effort, and impact, plus root-cause analysis, assumptions, open questions, and reproduction steps.
- Label reconciliation and stage transition: Applies status, effort, impact, and @mastra/core labels on GitHub issues and requests the governed Factory work-item stage transition as the terminal step.
- Use Case: A maintainer receives a vague bug report about agent memory in mastra-ai/mastra. The Skill traces the relevant code paths, reproduces the bug against the weather-agent repo, posts a triage comment classifying it as a high-severity regression, and advances the work item to Planning.
Quick Start
Triage the Factory work item for GitHub issue number 1234 and advance its stage.