resolve-pr-comments

Cluster GitHub PR comments and generate actionable.json for orchestration.

4|Updated Dec 14, 2025
One-click install
npx skills add https://github.com/LukasStrickler/ai-dev-atelier --skill resolve-pr-comments-lukasstrickler
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: resolve-pr-comments
Source: https://github.com/LukasStrickler/ai-dev-atelier/tree/main/content/skills/resolve-pr-comments
Command: npx skills add https://github.com/LukasStrickler/ai-dev-atelier --skill resolve-pr-comments-lukasstrickler

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires gh, jq, python3, git, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the handling of GitHub PR comments produced by automated reviewers (and human feedback) by clustering, prioritizing unresolved items, and coordinating deterministic agent workflows to fix, dismiss, or defer them at scale.

Core Features & Use Cases

  • Fetches and clusters PR comments into contextual groups to reduce cognitive load.
  • Spawns parameterized subagents for each actionable cluster to perform validation, fixes, or dismissals.
  • Generates actionable.json for orchestration and per-cluster Markdown files for subagents.
  • Supports an escalation path for deferred items with structured investigation templates.

Quick Start

  1. Run bash skills/resolve-pr-comments/scripts/pr-resolver.sh <PR_NUMBER> to fetch and cluster PR comments.
  2. Open .ada/data/pr-resolver/pr-<N>/actionable.json to view actionable clusters and plan work.
  3. For each actionable cluster, spawn a subagent using background_task with agent "pr-comment-reviewer" and the cluster file path.
  4. After all clusters complete, re-run the resolver to verify actionable_clusters == 0.

Frequently Asked Questions about resolve-pr-comments

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

FAQPage Schema
How do I automate GitHub PR comment resolution for large pull requests?

Automating GitHub PR comment resolution involves fetching reviewer and bot comments, grouping them by context, and coordinating subagents to fix or dismiss them at scale. This workflow generates actionable JSON files and markdown templates for structured triage.

What is the best way to cluster and triage unresolved PR comments?

Clustering and triaging unresolved PR comments is best handled by a script that fetches threads and groups them into contextual clusters, outputting actionable JSON and per-cluster markdown files. This parameterizes subagents to execute validation, fixes, or dismissals.

Do I need GitHub CLI and jq installed to orchestrate PR comment fixes?

Yes, you need GitHub CLI (gh), jq, python3, and git installed to orchestrate PR comment fixes. These dependencies support fetching comment data, processing JSON structures, and running the scripted utilities required for automated resolution workflows.

How does multi-agent orchestration work for automated PR comment resolution?

Multi-agent orchestration for PR comment resolution works by spawning parameterized subagents for each actionable comment cluster using a background task. Each subagent receives a cluster markdown file path, performs validation or fixes, and the main script is re-run to verify zero remaining clusters.

Can I defer GitHub PR comments instead of immediately fixing them?

Yes, you can defer GitHub PR comments instead of immediately fixing them. The orchestration workflow supports an escalation path for deferred items, providing structured investigation templates to manage postponed resolution tasks within the clustered triage process.

Why are my PR comments not clustering correctly during automated triage?

If PR comments are not clustering correctly during automated triage, ensure the prerequisite tools (gh, jq, python3) are properly configured and that the designated PR number contains unresolved threads. The script relies on fetching valid comment data to accurately generate contextual groups.