# Senior Code Review

> Use before finalizing a meaningful code change. Reviews the final diff for correctness, scope control, maintainability, validation gaps, safety risk, performance risk, accessibility risk, and subagent claims that still need verification.

- Skill: `arcanedge-ai/senior-code-review-2` (Agent Skill)
- Install (CLI): `npx skillmds@latest add arcanedge-ai/senior-code-review-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/arcanedge-ai/senior-code-review-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Web & Frontend
- Author: ArcanEdge-AI (https://skillmd.com/u/arcanedge-ai)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/arcanedge-ai/senior-code-review-2

---


# Senior Code Review Skill

Use this skill before finalizing meaningful code changes.

Review the final diff for:

- unrelated changes
- accidental formatting churn
- generated, vendored, compiled, or package-owned files
- missing or weak validation
- unused imports, variables, types, functions, or files caused by the change
- naming clarity
- consistency with existing patterns
- incomplete fixes that minimize the diff while leaving required behavior unresolved
- abstractions without a demonstrated boundary, invariant, meaningful duplication, or variability
- unnecessary change amplification across unrelated components
- material technical debt without its scope, rationale, and follow-up condition
- speculative configurability
- behavior changes beyond the request
- API compatibility
- migration risk
- safety risk
- performance risk
- accessibility regressions
- subagent claims that were not independently verified
- finite-manifest nodes, root permits, or total-budget use that were not reconciled
- child execution that did not select a verified Luna/max profile or explicit `gpt-5.6-luna`/`max` child-execution settings, or permissions, scope, authority, or workspace expansion beyond the parent assignment
- progress or task-reporting messages that set `model`, `reasoning_effort`, `thinking`, or analogous destination-setting overrides, or altered a parent or peer task
- task-created auxiliary worktrees without integration evidence and a verified `removed` or exact-blocker `preserved` disposition

Ask:

```text
Would I approve this in code review?
```

If not, fix the issue or report the remaining risk clearly.

Final report format:

```text
Summary:
- Changed X to do Y.

Verification:
- Ran: [command or check]
- Result: [passed/failed/not run, with reason]

Notes:
- [Assumptions, unrelated issues, subagent usage, follow-ups, or risk notes if any]
```