skills.rest
PostHog
PostHogOFFICIAL
Explore all skills by this author →
Website𝕏@PostHogGitHub

review-hog-blind-spots-general

Detects high-value PR issues missed by all specialist review perspectives.

713|118|Updated Aug 11, 2020
Search Tags:#static analysis#edge cases#blind spots#PostHog#review automation#Code review#Pull request
One-click install
npx skills add https://github.com/PostHog/posthog-foss --skill review-hog-blind-spots-general
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-hog-blind-spots-general
Source: https://github.com/PostHog/posthog-foss/tree/main/products/review_hog/skills/review-hog-blind-spots-general
Command: npx skills add https://github.com/PostHog/posthog-foss --skill review-hog-blind-spots-general

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Specialist code review perspectives each focus on one lens, so issues falling between their scopes slip through. This Skill runs a final sweep over a PR chunk, conditioned on what every perspective already found, to catch the real problems none of them covered.

Core Features & Use Cases

  • Gap-Conditioned Review: Studies the findings from all enabled review perspectives first, then hunts only in the territory they did not examine.
  • Cross-Specialty Scope: Reports real issues from any lens — untested edge cases, error paths, unhandled inputs, cross-file interactions — as long as no perspective already raised them.
  • Empty-List Discipline: Returns an empty issues list when the perspectives were thorough, avoiding duplicate or padded findings.
  • Use Case: After security, performance, and correctness perspectives each review a PR chunk, run this sweep to surface a missed failure mode in an unhandled input path that none of them inspected.

Quick Start

Run the blind-spot check on this PR chunk after the other review perspectives finish, and report only the new issues they all missed.

Frequently Asked Questions about review-hog-blind-spots-general

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

FAQPage Schema
How do I catch code review issues that specialist reviewers miss?▼

Run a final blind-spot sweep after all specialist perspectives finish reviewing a chunk. It conditions on their actual findings and hunts only in areas they did not examine, such as untested edge cases, error paths, and cross-file interactions.

What is a blind-spot check in automated PR review?▼

A blind-spot check is the last pass of a multi-perspective PR review. It reads what each perspective found and reports only genuinely new, concrete problems with a nameable trigger and consequence anchored to the chunk's changes.

Does the blind-spot check duplicate findings from other reviewers?▼

No. It is explicitly conditioned on the covered findings and inline PR comments, and it must not re-report, restate, or reword anything already raised. Only genuinely new problems are reported.

What happens when reviewers were thorough and nothing was missed?▼

The check returns an empty issues list, which is treated as a valid and good outcome. Padding the results with low-value findings is explicitly discouraged.

When should the blind-spot sweep run in a review pipeline?▼

It runs last, after every enabled review perspective has reviewed the chunk. It needs the list of which perspectives ran and what they found so it can target the gaps instead of re-walking covered ground.

Related Skills

Explore More Software Engineering Skills→
hawkyrehawkyre
30

review-large-pr

Partition pull requests into chunks and run parallel specialist audits.

Community
Advanced
TheFVGTheFVG

babysit

Review and resolve PR comments, merge conflicts, and CI issues.

Official
Intermediate
wendeus0wendeus0

test-coverage-report

Analyze coverage reports and flag modules below the AGENTS.md threshold.

Community
Intermediate
openaiopenai
5.3k

security-scan

Runs a single-pass security audit of a repository or scoped path and generates a findings report.

Official
Advanced