diff-review

diff-review is a command for coding agents from huurray/fe-review-agents. It costs 89 tokens per session (1,624 once invoked), scanned A, original, MIT.

A command that reviews frontend changes in a Git diff using several specialized reviewers and one synthesizer. A Git diff shows what code was added, changed, or removed.

In plain words
What is it for?
Reviewing staged, unstaged, branch, or commit-range changes, filtering frontend files, and producing a severity-ordered report in Korean or English.
Why use it?
It focuses review on the actual changes and combines findings from performance, quality, bug, TypeScript, accessibility, and security checks.

Command

Part of the fe-review-agents plugin — 2 commands, 7 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/huurray/fe-review-agents/diff-review
Clone the repo
git clone --depth 1 https://github.com/huurray/fe-review-agents

Or install fe-review-agents, the plugin that ships this one along with the rest of its 2 commands, 7 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for diff-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/huurray/fe-review-agents/diff-review.svg)](https://agentmods.dev/commands/huurray/fe-review-agents/diff-review)
Your own site
<a href="https://agentmods.dev/commands/huurray/fe-review-agents/diff-review"><img src="https://agentmods.dev/badge/commands/huurray/fe-review-agents/diff-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 89 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,624 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00089 $0.01624
Opus 5 $0.00044 $0.00812
Sonnet 5 $0.00018 $0.00325
Haiku 4.5 $0.00009 $0.00162

Measured 5d ago against content hash 184ad38628d9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

diff-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/diff-review.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

git diff 기반 리뷰 워크플로우. 여러 reviewer + synthesizer.

단계 0 — Git 확인 + 인자 파싱 + diff 추출

먼저 Bashgit rev-parse --is-inside-work-tree를 실행해 git 레포인지 확인. 아니면 "git 레포가 아닙니다" 알리고 종료.

$ARGUMENTS에서 다음 두 옵션 토큰을 먼저 추출:

  • lang=ko|en (기본 ko)
  • severity_min=LOW|MED|HIGH|CRITICAL (기본 LOW. 대소문자 무시 — 내부적으로 대문자로 정규화)

각 토큰의 값이 위 허용 목록에 없으면 default로 fallback하고 사용자에게 한 줄 경고를 그대로 출력 (예: severity_min=foo`는 무효 — `LOW`로 진행). 두 토큰을 제거한 나머지를 scope로 해석:

scope 토큰 실행 명령
비어있음 또는 staged git diff --cached --unified=3
unstaged git diff --unified=3
branch:<name> git diff <name>...HEAD --unified=3
range:<a>..<b> git diff <a>..<b> --unified=3

위 형식에 안 맞으면 사용자에게 사용법 알리고 종료.

auto 모드 보강: 만약 사용자가 scope 인자 없이 호출했고 staged diff가 비어있으면, unstaged로 한 번 더 시도.

단계 1 — 프론트엔드 파일만 필터링

diff에서 다음 확장자만 남기고 나머지 파일 hunk는 제거: .ts .tsx .js .jsx .mjs .cjs .vue .svelte .html .css .scss

다음은 제외: .d.ts, dist/**, build/**, .next/**, node_modules/**, *.test.*, *.spec.*, __tests__/**, *.stories.*

  • 필터 후 비어있으면: "리뷰할 프론트엔드 변경 없음" 알리고 종료.
  • 필터 후 라인 수 2,000 초과면: "diff가 너무 큽니다. scope를 좁혀주세요 (예: branch:main, range:HEAD~1..HEAD)" 안내하고 종료.

단계 1.5 — 필터된 diff를 임시 파일에 저장

병렬 dispatch를 위해 필터된 diff 텍스트를 /tmp/fe-review-diff.txt에 저장합니다. 두 단계로:

  1. 먼저 Bashrm -f /tmp/fe-review-diff.txt 실행. 기존 파일이 있으면 (이전 세션 잔존물 등) 삭제, 없으면 no-op (-f 플래그가 에러 억제). 이 단계로 Write 도구의 "exists → Read first" 요건 회피 + stale 파일 권한 충돌 방지.
  2. 그 다음 Write 도구로 필터된 diff 내용을 /tmp/fe-review-diff.txt에 저장.

이렇게 하면 단계 2의 dispatch prompt가 작아져서 (path만 전달) 모든 Agent 호출이 하나의 어시스턴트 메시지에 한꺼번에 담겨 실제로 병렬 실행됩니다. (diff 텍스트를 prompt에 인라인하면 출력이 거대해져 모델이 dispatch를 스스로 나눠 → 직렬 실행으로 바뀝니다.)

각 reviewer는 자기 sub-agent 컨텍스트에서 Read로 이 파일을 한 번씩 읽습니다 (병렬이라 wall-time에 영향 미미).

단계 2 — 모든 reviewer를 병렬로 호출

반드시 하나의 어시스턴트 메시지 안에, 아래 나열된 모든 Agent 도구 호출을 한꺼번에 보내야 합니다. (순차 호출 금지)

각 reviewer에 다음 형식의 작은 prompt를 전달 (<LANG>은 단계 0에서 결정한 언어):

다음 경로의 git diff 파일을 [관점] 관점에서 리뷰하세요.
(diff 모드 — Read로 파일을 읽고 그 안의 diff 텍스트만 분석. 일반 source code 리뷰가 아님)

lang=<LANG>

DIFF 파일 경로: /tmp/fe-review-diff.txt

Read the full file on GitHub · 111 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 111 lines · 89 tokens per session scan A 184ad38628d9

Subscribe to this mod's changes

diff-review is a command published in the GitHub repository huurray/fe-review-agents (5 stars, last pushed 3mo ago), licensed MIT. It adds 89 tokens to every session and 1,624 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.