skills.rest
generic-automation-and-it
generic-automation-and-itOFFICIAL
Explore all skills by this author →
GitHub

ai-review

Parses AI pull request reviews and applies fix or skip decisions on GitHub.

Updated Jul 23, 2026
Search Tags:#automation#code review#pull request#github#gh cli#triage#copilot
One-click install
npx skills add https://github.com/generic-automation-and-it/smooth-ai-stockanalysis --skill ai-review-generic-automation-and-it
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-review
Source: https://github.com/generic-automation-and-it/smooth-ai-stockanalysis/tree/main/.agents/skills/ai-review
Command: npx skills add https://github.com/generic-automation-and-it/smooth-ai-stockanalysis --skill ai-review-generic-automation-and-it

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? AI-generated PR reviews produce lists of findings that someone must manually triage, fix, and respond to across scattered review threads. This Skill automates that loop: it parses the review, recommends fix or skip per issue, applies the chosen fixes, and routes the results back to the right place on the pull request. ## Core Features & Use Cases - Analyse mode: Fetches the AI review for a PR, extracts each issue with priority and reasoning, and outputs a recommendation table without changing any code. - Execute mode: Applies per-issue decisions (e.g. 1=fix 2=skip), commits fixes, and enforces a mandatory empty re-verification commit when critical or high findings exist. - Source-aware routing: Auto-detects GitHub Copilot reviews versus other AI reviewers; Copilot results are posted as thread replies and resolutions, while other reviews append notes to the PR description. - Use Case: After a Copilot review flags five issues on PR 48, run analyse to get recommendations, then execute with your decisions so each thread gets a reply, is resolved, and a summary comment is posted. ## Quick Start Ask the agent to analyse the AI review on pull request 48 and recommend which findings to fix or skip.

Frequently Asked Questions about ai-review

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

FAQPage Schema
How do I process a GitHub Copilot PR review automatically?▼

Run the skill in analyse mode with the PR number to get a fix/skip recommendation table, then run execute mode with decisions like 1=fix 2=skip. It replies to each Copilot review thread, resolves it, and posts a summary comment on the PR.

How does the skill detect Copilot versus other AI reviewers?▼

It runs copilot-review.sh detect, which scans the PR's reviews and inline comments for the Copilot reviewer bot logins. You can override detection with --source=copilot or --source=other.

Can I use this with non-Copilot AI code reviews?▼

Yes. For reviews from other sources such as OpenCode or Gemini, the skill appends the fix/skip summary table to the PR description's AI Review Notes section instead of touching review threads, and it never overwrites existing notes.

Why does execute mode create an empty commit after fixing review findings?▼

When a review contains any Critical or High finding, the skill makes a mandatory empty commit to re-trigger a full review run that re-verifies those findings. Reviews with only medium or low findings skip this commit.

What tools are required to run the AI review workflow?▼

The workflow requires the GitHub CLI (gh) for REST and GraphQL calls and jq for paginated review thread fetching. All GitHub operations are handled by the bundled copilot-review.sh script.

Related Skills

Explore More Software Engineering Skills→
pydanticpydantic
8.1k

fix-pr-comments

Reads, resolves, and replies to bot reviewer comments on GitHub pull requests.

Official
Intermediate
pydanticpydantic
19.2k

address-feedback

Resolve unresolved GitHub pull request review comments on the current branch.

Official
Advanced
ghufronbagaskaraghufronbagaskara

requesting-code-review

Automate pre-commit code verification with static security analysis, regression testing, and subagent review.

Community
Advanced
NVIDIANVIDIA
49

sop-ddm-finetuning

Fine-tune DDM-Net temporal boundary detectors on annotated SOP video datasets.

Official
Advanced