skills.rest
wilsonfaustino
wilsonfaustinoCOMMUNITY
Explore all skills by this author →
GitHub

staff-review

Audit pull request comments and surface high-confidence code issues.

2|Updated Apr 11, 2026
Search Tags:#code review#github#pr review#staff engineer#pull request validation#false positive filtering#review auditing
One-click install
npx skills add https://github.com/wilsonfaustino/ai-tools --skill staff-review-wilsonfaustino
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: staff-review
Source: https://github.com/wilsonfaustino/ai-tools/tree/main/skills/staff-review
Command: npx skills add https://github.com/wilsonfaustino/ai-tools --skill staff-review-wilsonfaustino

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Standard PR reviews often miss critical code issues, leave unvalidated or outdated reviewer feedback unaddressed, and generate excessive false positives from automated tools that waste engineering time during code triage. This Skill delivers high-confidence, staff-engineer-level PR reviews that eliminate noise and ensure only actionable feedback is surfaced.

Core Features & Use Cases

  • Parallel Multi-Source Review: Fans out to pr-review-toolkit and pr-review-local in parallel to combine automated code analysis, local subagent checks for security, regression, and performance issues, and silent failure detection.
  • Review Comment Auditing: Fetches and audits all existing automated and human PR review comments, flagging outdated, resolved, or incorrect feedback so authors don't waste time on already-addressed concerns.
  • Judge Layer for Precision: Runs a dedicated judge subagent to deduplicate overlapping findings, filter false positives, and assign confidence scores, so only high-signal issues are presented to the user.
  • Use Case: Ideal for engineering teams merging complex feature branches or bug fixes, where missing a critical regression or security gap could cause production incidents, and where multiple reviewers have left overlapping or conflicting feedback.

Quick Start

Use the staff-review skill to run a read-only, thorough review of the current pull request, including validation of all existing reviewer comments and false-positive filtering.

Frequently Asked Questions about staff-review

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

FAQPage Schema
How do I eliminate false positives from automated PR reviews?▼

To eliminate false positives from automated PR reviews, a judge subagent deduplicates overlapping findings, filters incorrect feedback, and assigns confidence scores so only high-signal, actionable code issues are surfaced to the author.

How do I audit existing pull request comments for outdated feedback?▼

Auditing existing pull request comments involves fetching all automated and human reviewer feedback, then flagging outdated, resolved, or incorrect entries so authors avoid wasting time on already-addressed concerns during code review.

What is the best way to detect silent failures and security gaps before merging?▼

Detecting silent failures and security gaps before merging requires parallel multi-source analysis that combines automated tool output with local subagent checks to thoroughly validate complex code changes for high-confidence issue detection.

Does staff-level PR review require GitHub CLI authentication?▼

Staff-level PR review requires GitHub CLI authentication and installed pr-review-toolkit and optional pr-review-local dependencies to fetch repository data and run comprehensive code analysis.

Can a pull request validation tool modify my code during review?▼

Pull request validation tooling can enforce read-only operation to prevent unintended PR mutations during review, ensuring the codebase remains completely untouched while analyzing security gaps and performance regressions.

When should I use staff-engineer-level code review for my pull requests?▼

Staff-engineer-level code review is ideal when merging complex feature branches or bug fixes where missing a critical regression, security gap, or unaddressed reviewer feedback could cause production incidents.

Related Skills

Explore More Software Engineering Skills→
granietgraniet
247

github-code-review

Analyze local git diffs and pull request changes for bugs and security issues.

Community
Intermediate
ZaxbyHubZaxbyHub
437

swarm-pr-review

Review pull requests with multi-lane validation and critic challenge.

Official
Advanced
matteobortolazzomatteobortolazzo
1

review

Automate code reviews for security, quality, and silent-failure detection.

Community
Advanced
dreadnodedreadnode
11

vuln-critic

Filter low-quality and false positive vulnerability findings from scanners and agent output.

Official
Intermediate