Skip to main content
skill CX Codex

review-hog-validation-criteria — Claude Code skill: setup and features

review-hog-validation-criteria serves as a final PR review gate, filtering candidate findings to keep real defects while discarding speculative noise.

no license 187 installs ★ 101 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add PostHog/ai-plugin --skill review-hog-validation-criteria -a claude-code
Open on GitHub

What review-hog-validation-criteria does

The skill functions as a decision gate for pull request reviews. Instead of inspecting code from scratch or inventing fresh problems, it assesses candidate issues raised by previous checkers. It inspects callers, validation logic, and surrounding context in the live tree before deciding whether each candidate warrants surfacing to the author.

Evaluation relies on a strict precision-over-recall threshold. Findings remain valid only when they describe concrete defects: logic bugs, authentication or injection vulnerabilities, data loss, API contract breaks, and high-load performance drops like N+1 queries. Pure cosmetic critiques, naming debates, defensive guards against impossible states, and hypothetical edge cases get thrown out.

When a candidate lacks both an observable execution path and an explicit consequence, the skill drops it immediately.

What it helps with

  • filter speculative suggestions and cosmetic remarks out of code reviews
  • confirm whether a reported defect is truly reachable through existing callers
  • suppress redundant defensive checks already handled by types or frameworks
  • surface only actionable flaws involving correctness, security, or API contracts

Example prompts for the agent

  • Filter these candidate PR review findings using review-hog-validation-criteria
  • Verify if this flagged performance defect is reachable in the codebase before reporting it

How to install review-hog-validation-criteria

1Skills CLI (Vercel)

npx skills add PostHog/ai-plugin --skill review-hog-validation-criteria -a claude-code

2Manually: put the files in place

~/.claude/skills/review-hog-validation-criteria/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/review-hog-validation-criteria/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about review-hog-validation-criteria

How to install review-hog-validation-criteria in Claude Code?

Run npx skills add PostHog/ai-plugin --skill review-hog-validation-criteria -a claude-code. The file will be placed in ~/.claude/skills/review-hog-validation-criteria/SKILL.md.

What does review-hog-validation-criteria do?

It evaluates pre-flagged review candidates against the live codebase, setting is_valid to true only for verifiable bugs, security flaws, or breaking changes.

How does the skill handle uncertain findings?

It applies a precision-over-recall rule: whenever evidence fails to confirm real impact, the issue is dropped rather than surfaced.

Similar skills

All skills for Code review and refactoring →

From the same set82

The whole set →
analyzing-expensive-users analyzing-expensive-users — a skill for Claude Code, Codex, Cursor and other agents that finds the top spenders in PostHog AI observability and explains what drives their costs. CX Codex +1 192installs turning-engineering-analytics-into-insights turning-engineering-analytics-into-insights — a skill for Claude Code and Codex: converts PR and CI metrics into saved PostHog insights, dashboards, and automated subscriptions. Triggers when visualizing engineering team performance. CX Codex +1 189installs investigating-logs The investigating-logs Claude Code skill investigates logs in a PostHog project, checks service health and traces error spikes. It runs for deploy checks, incident triage, change analysis, or a search around one trace or run. CX Codex +1 188installs investigating-ci-failures The investigating-ci-failures Claude Code skill traces an individual CI failure to its likely cause. It separates PR-specific problems from default-branch breaks and flaky tests. CX Codex +1 188installs review-hog-blind-spots-general The review-hog-blind-spots-general skill looks for concrete problems that earlier reviewers missed in a PR chunk. Use it as a final pass after other review perspectives have reported their findings. CX Codex +1 187installs review-hog-authoring review-hog-authoring — a guide skill for writing custom PostHog Review pieces: extra PR perspectives, your own blind-spot sweep, and the bars that decide what gets published or fixed. CX Codex +1 187installs review-hog-perspective-contracts-security review-hog-perspective-contracts-security is a Claude Code skill that checks code changes for security flaws and API compatibility breaks. It reports contract and security findings during PR review. CX Codex +1 187installs review-hog-perspective-performance-reliability The review-hog-perspective-performance-reliability skill reviews code for performance and reliability concerns, from database access to error recovery. Use it when checking a change before it ships. CX Codex +1 187installs

Project sponsors

Companies that keep this analytics open