issue-triage

Detect duplicate, stale, and misaligned Linear issues via multi-agent consensus.

1|Updated Aug 16, 2025
One-click install
npx skills add https://github.com/RB-chrismandich/Manifest --skill issue-triage-rb-chrismandich
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: issue-triage
Source: https://github.com/RB-chrismandich/Manifest/tree/main/configs/claude/skills/issue-triage
Command: npx skills add https://github.com/RB-chrismandich/Manifest --skill issue-triage-rb-chrismandich

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This backlog triage solution automates detection of duplicates, stale issues, and priority misalignments in Linear, helping product and engineering teams maintain a healthy backlog.

Core Features & Use Cases

  • Detect duplicates across Linear issues using semantic similarity and cross-team validation.
  • Identify stale/obsolete issues (inactive for long periods, unassigned, or with zero labels) and surface safe-to-close candidates.
  • Validate and surface misalignments between current priorities and computed impact/urgency, enabling data-driven backlog updates.
  • Use parallel agents for complex decisions and generate actionable recommendations with confidence scores.
  • Produce a concise triage report and an audit trail to support governance and rework.

Quick Start

Run /issue-triage with your desired flags to start triaging your Linear backlog and optionally act on high-confidence outcomes.

Frequently Asked Questions about issue-triage

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

FAQPage Schema
How do I automate Linear issue triage and backlog cleanup?

Automate Linear backlog cleanup by running the triage process to detect duplicate issues, flag stale items, and validate priority alignment using multi-agent consensus. It generates actionable recommendations with confidence scores and an audit trail.

What is the best way to find duplicate Linear issues across teams?

Finding duplicate Linear issues across teams is handled through semantic similarity analysis and cross-team validation. Parallel agents evaluate the backlog to identify overlapping items and generate recommendations with confidence scores.

How do I identify stale Linear issues that are safe to close?

Identify stale Linear issues by detecting items that are inactive for long periods, unassigned, or have zero labels. The triage process surfaces safe-to-close candidates and can auto-close them when approved via safety rules.

Do I need a Linear MCP setup to run backlog triage?

A Linear MCP setup is required to run backlog triage, along with API keys and tooling availability. These prerequisites enable the parallel agents and local rule engine to access your issues and coordinate triage actions.

Can I run a dry-run before auto-closing stale Linear issues?

Dry-runs are supported to preview triage outcomes before executing live actions. Safety rules ensure that auto-closing stale issues only occurs when explicitly approved, producing an audit trail for governance and rework.

How does multi-agent consensus validate Linear priority alignment?

Multi-agent consensus validates Linear priority alignment by comparing current priorities against computed impact and urgency. Parallel agents coordinate through a local rule engine to surface misalignments and enable data-driven backlog updates.