# Code Reviewer > Review code for bugs, security flaws, maintainability risks, and missing test coverage. Use when evaluating a diff, pull request, or source file before merge. Use when this capability is needed. - Skill: `tomevault-io/code-reviewer-31` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/code-reviewer-31` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/code-reviewer-31/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/code-reviewer-31 --- # Code Reviewer ## Review Workflow 1. Identify language/framework and intended behavior. 2. Scan for correctness and regression risk first. 3. Check security-sensitive paths and input handling. 4. Evaluate maintainability and readability tradeoffs. 5. Identify missing tests and suggest targeted additions. ## Output Format - Findings ordered by severity. - File + line reference for each issue. - Concrete fix suggestion. - Short summary of residual risks. ## Severity Guide - Critical: exploitability, data loss, auth bypass. - High: likely production breakage. - Medium: correctness or maintainability issue with clear impact. - Low: polish and consistency issues. ## Edge Cases - If context is incomplete, state assumptions explicitly. - If no issues are found, state that clearly and highlight test gaps. ## Reference Use `references/checklist.md` for a consistent scan order. --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/danaidev) — claim your Tome and manage your conversions.