swarm-pr-feedback

Resolve pull request feedback with verified source evidence and closure summaries.

437|45|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/ZaxbyHub/opencode-swarm --skill swarm-pr-feedback
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: swarm-pr-feedback
Source: https://github.com/ZaxbyHub/opencode-swarm/tree/main/.opencode/skills/swarm-pr-feedback
Command: npx skills add https://github.com/ZaxbyHub/opencode-swarm --skill swarm-pr-feedback

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

PR feedback is scattered across GitHub reviews, CI failures, bot comments, and pasted notes, making it easy to miss findings, mishandle false claims, or ship code with unreviewed issues. This Skill eliminates that risk by ingesting every feedback source, verifying each claim with source evidence, and ensuring every item is explicitly addressed before the PR can merge.

Core Features & Use Cases

  • Unified Feedback Ledger: Ingests all PR feedback surfaces (GitHub review threads, CI/check failures, bot review findings, pasted reviewer comments, merge conflicts, stale branch state) into a single normalized ledger with unique IDs and provenance tracking.
  • Verified Claim Triage: Classifies every feedback item as confirmed, disproved, pre-existing, or needing user input using async verification lanes and source code evidence, avoiding false fixes for incorrect bot or reviewer claims.
  • Multi-Round Bot Review Support: Handles iterative automated bot reviews that post new findings after every push, carrying forward unresolved items across rounds instead of re-triaging from scratch, and applying a 3-strikes rule to stop unproductive debate over repeated false claims.
  • Use Case: For a PR that has received 3 rounds of automated bot reviews, 2 human reviewer threads, and a failing CI check, use this Skill to track every finding, verify each claim against the current branch, fix confirmed issues, and generate a closure summary showing the status of every item.

Quick Start

Use the swarm-pr-feedback skill to ingest all existing pull request feedback, verify each claim with source evidence, fix confirmed issues, and generate a complete closure ledger for the PR.

Frequently Asked Questions about swarm-pr-feedback

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

FAQPage Schema
How do I resolve all pull request feedback including CI failures and bot review findings before merge?

Resolve all pull request feedback by ingesting CI failures, bot findings, and review comments into a unified ledger, verifying each claim with source code evidence, fixing confirmed issues, and generating a complete closure summary.

What is the best way to track and verify multi-round bot review feedback on a PR?

Tracking multi-round bot review feedback requires normalizing all iterative findings into a ledger that carries forward unresolved items across pushes, avoiding re-triaging, and applying a 3-strikes rule to halt unproductive debate over repeated false claims.

How do I handle incorrect bot or reviewer claims without applying false fixes to my branch?

Handle incorrect bot or reviewer claims by classifying every feedback item as confirmed, disproved, or pre-existing using async verification lanes that check source code evidence before any fix is applied.

Can I track merge conflicts and stale branch state alongside human reviewer comments in a single view?

Yes, you can track merge conflicts, stale branch state, and human reviewer comments in a single normalized ledger that assigns unique IDs and provenance tracking to every PR feedback item.

How do I generate a closure summary mapping every PR feedback item to its final outcome?

Generate a closure summary by enforcing a mandatory feedback ledger that tracks every item's provenance and status, ensuring every piece of PR feedback is explicitly addressed and mapped to its final resolution before merging.

What should I do when automated bot reviews repeatedly post the same false findings across multiple pushes?

Manage repeatedly false bot findings across pushes by carrying forward unresolved items in a feedback ledger instead of re-triaging, and apply a 3-strikes rule to stop unproductive automated debate over disproved claims.