skills.rest
oncokb
oncokbOFFICIAL
Explore all skills by this author →
Website𝕏@OncoKBGitHub

oncokb-ai-code-review

Runs an AI-assisted PR review validating issue linkage, quality gates, scope, and review comments.

Updated Aug 31, 2026
Search Tags:#testing#code review#pull request#accessibility#github#lint#ai review
One-click install
npx skills add https://github.com/oncokb/oncokb-skills --skill oncokb-ai-code-review-oncokb
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: oncokb-ai-code-review
Source: https://github.com/oncokb/oncokb-skills/tree/main/skills/oncokb-ai-code-review
Command: npx skills add https://github.com/oncokb/oncokb-skills --skill oncokb-ai-code-review-oncokb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual pull request reviews often miss database risks, logging issues, accessibility problems, and test gaps, and reviewers lose track of which checks were actually run. This Skill enforces an auditable, end-to-end AI review workflow on an existing PR. ## Core Features & Use Cases - Structured review chain: Orchestrates issue linkage, local format/lint/test gates, scope checks, and specialized reviews for database access, logging, HTML accessibility, Datastar, code comments, and test coverage. - Scope governance: Measures diff size against soft limits of 500 changed source lines and 20 changed files, recommending a PR stack split when exceeded. - Auditable PR comments: Posts actionable review comments prefixed with [ai-generated] and reports every command run with its outcome. - Use Case: A developer opens a PR touching database queries and UI markup; the Skill confirms the linked GitHub issue, runs lint and tests, flags an N+1 query and a missing form label, and posts [ai-generated] comments on the PR. ## Quick Start Ask the agent to run an AI code review on your pull request URL using the oncokb-ai-code-review workflow.

Frequently Asked Questions about oncokb-ai-code-review

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

FAQPage Schema
How do I run an AI code review on a GitHub pull request?▼

Provide the related GitHub issue and the PR URL, then the workflow runs local format, lint, and test gates before posting review comments. All automated comments are prefixed with [ai-generated] and tied to specific files or concerns.

What checks does an AI PR review workflow include?▼

The workflow covers issue linkage, formatting, linting, tests, diff scope limits, database access risks like N+1 queries and SQL injection, logging level correctness, HTML accessibility, Datastar usage, code comment quality, and test coverage adequacy.

Can this skill create or update a pull request?▼

No, it only reviews existing pull requests. If no PR exists yet, the workflow stops and hands off to the oncokb-make-pull-request skill before any review work begins.

What happens when a pull request is too large to review?▼

Diffs exceeding 500 changed source lines or 20 changed source files trigger a recommendation to split the work into a PR stack. The reviewer receives either a manageable-scope confirmation or a concrete split plan.

Why does the review stop before posting PR comments?▼

The review halts when formatting, lint, or test commands fail, when no issue reference exists, or when the PR URL is missing. The exact blocker and the next required action are reported instead of proceeding.

Related Skills

Explore More Software Engineering Skills→
HomericIntelligenceHomericIntelligence
18

gh-create-pr-linked

Create GitHub pull requests linked to specific issues via the GitHub CLI.

Official
Intermediate
zizzfizzixzizzfizzix

sync-merged-pr-issues

Reconcile merged pull requests with GitHub issues and post comments.

Community
Intermediate
stacklokstacklok
2.1k

code-review-assist

Summarizes pull request diffs, surfaces review concerns, and assesses test coverage for human reviewers.

Official
Advanced
khoavu882khoavu882

do-implement

Implements code changes directly from descriptions, review findings, or named tasks without requiring planning artifacts.

Community
Advanced