resolve-ai-reviews

Triage and resolve AI bot review comments on GitHub pull requests.

Updated Dec 25, 2021
One-click install
npx skills add https://github.com/kotahashihama/dotfiles --skill resolve-ai-reviews-kotahashihama
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: resolve-ai-reviews
Source: https://github.com/kotahashihama/dotfiles/tree/main/home/.claude/skills/resolve-ai-reviews
Command: npx skills add https://github.com/kotahashihama/dotfiles --skill resolve-ai-reviews-kotahashihama

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI code review bots (like GitHub Actions reviewers) leave many inline comments on pull requests, and blindly applying every suggestion can overwrite decisions already agreed upon with the user. This Skill fetches bot review comments, cross-checks each one against the session context (explicit user instructions, PR body notes, prior agreements, memory files), and only applies context-consistent fixes before replying and resolving threads. ## Core Features & Use Cases - Context-aware triage: Builds a session context snapshot and classifies each bot comment as fix, decline-with-reason, or ask-the-user, using parallel subagents when there are 3 or more comments. - Full GitHub workflow: Applies fixes, runs quality checks (format/lint/test), commits and pushes, replies with the commit hash or a rejection reason plus a generated-by note, adds +1 reactions, and resolves review threads via GraphQL. - Use Case: After opening a PR, the AI reviewer leaves 8 comments. Run this Skill to automatically fix the valid ones, politely decline the ones that conflict with documented design decisions, and resolve every replied thread. ## Quick Start Ask the AI to handle the AI review comments on the current pull request, for example by saying "resolve the AI review comments on this PR".

Frequently Asked Questions about resolve-ai-reviews

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

FAQPage Schema
How do I respond to AI code review comments on a GitHub pull request?

Fetch inline review comments with the GitHub API, triage each by priority prefix like [must] or [nits], apply valid fixes, then reply with the commit hash and resolve the conversation thread. This Skill automates that entire loop using the gh CLI.

How to resolve review threads on GitHub pull requests via API?

GitHub thread resolution requires the GraphQL API, not REST. Query reviewThreads to get thread IDs mapped to comment databaseIds, then run the resolveReviewThread mutation with each thread ID and confirm isResolved returns true.

Should I apply every AI code review suggestion automatically?

No. Suggestions can conflict with explicit user instructions, documented design decisions in the PR body, or recorded preferences. This Skill cross-checks each comment against that context and asks the user before acting on conflicting or ambiguous points.

Does this work on pull requests created by other people?

No. The Skill checks the PR author first and stops if the pull request belongs to someone else, since replying, resolving threads, and pushing would leave traces on another person's PR.

What tools are required to automate PR review responses?

You need the GitHub CLI (gh) for REST and GraphQL API calls, git for committing and pushing fixes, and your project's quality check commands such as format, lint, and test runners to validate changes before pushing.