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.
npx skills add sh5623/fe-rail --skill fe-reviewgit clone --depth 1 https://github.com/sh5623/fe-railWrote 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.
[](https://agentmods.dev/skills/sh5623/fe-rail/fe-review)<a href="https://agentmods.dev/skills/sh5623/fe-rail/fe-review"><img src="https://agentmods.dev/badge/skills/sh5623/fe-rail/fe-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sh5623/fe-rail/fe-review"><img src="https://agentmods.dev/badge/skills/sh5623/fe-rail/fe-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00046 | $0.00896 |
| Opus 5 | $0.00023 | $0.00448 |
| Sonnet 5 | $0.00009 | $0.00179 |
| Haiku 4.5 | $0.00005 | $0.00090 |
Grade A, and why
fe-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.
What it actually says
FE Review 스킬
구현된 코드를 타입·성능·a11y·품질 4개 축으로 병렬 검토합니다.
When to Use
- 구현 완료 직후
- 커밋 전 마지막 품질 게이트
- PR 생성 전
Instructions
Phase 1 — 에이전트 위임
리뷰는 직접 수행하지 않고 전담 에이전트에 위임한다:
| 에이전트 | 역할 | 위임 조건 |
|---|---|---|
fe-reviewer |
타입·성능·a11y·품질 4축 기본 리뷰 | 항상 |
fe-a11y-auditor |
WCAG AA 정밀 감사 | fe-reviewer에서 접근성(a11y) BLOCK/WARN 발생 시 |
fe-perf-auditor |
번들·Image·Font·Tailwind purge 정밀 감사 (Next.js / Vite SPA 공통) | fe-reviewer에서 성능 BLOCK/WARN 발생 시 |
fe-test-runner |
테스트 실행·실패 분류 | 항상 (스택트레이스는 에이전트 내부에서 처리) |
fe-refactor-advisor |
리팩토링 방향 분석 | 명시적 요청 시 |
모든 에이전트는 READ-ONLY — 메인 세션이 결과를 받아 수정 여부를 결정한다.
Phase 2 — 리뷰 결과 보고 (합산 판정)
fe-reviewer 의 출력을 본문으로 전달하되, "커밋 준비 완료" 판정은 메인이 합산해서 내린다 — reviewer BLOCK 0
그리고 fe-test-runner exit 0 그리고 추가 감사(a11y·perf) BLOCK 0 일 때만. 테스트가 실패했는데 reviewer
BLOCK 이 0 이라고 "준비 완료" 로 적지 않는다. 요약 줄에 각 소스의 결과를 나란히 적는다:
리뷰 BLOCK 0 · 테스트 exit 0 (42/42) · a11y BLOCK 0 → 커밋 준비 완료 / 리뷰 BLOCK 0 · 테스트 exit 1 (2 실패) → 수정 후 재검토.
위임 컨텍스트에는 변경 파일 목록(신규 파일 포함) 을 넘긴다 — 에이전트는 tracked diff ∪ untracked 와 교차 확인한다.
그리고 docs/framework-rules.md 의 절대경로(이 스킬이 로드될 때 주어진 base directory 기준 ../../docs/framework-rules.md)를 함께 넘긴다 — fe-reviewer·fe-perf-auditor 는 그 파일의 «공통 규칙» 절 + 해당 프레임워크 절을 판정 근거로 쓴다(소비자 프로젝트 CLAUDE.md 가 우선). 에이전트는 플러그인 트리 위치를 모르므로 경로 없이 위임하면 내장 기준만으로 판정하고 그 사실을 보고 첫 줄에 적는다. 모노레포면 ../../docs/monorepo.md 경로도 넘긴다.
형식:
## 코드 리뷰 결과
### BLOCK (커밋 전 반드시 수정)
- [축] file:line — 문제
- Before: ...
- After: ...
### WARN (권장 수정)
- [축] file:line — 문제
### INFO (참고)
- [축] file:line — 내용
---
요약: BLOCK N개 / WARN N개 / INFO N개 · 테스트 exit N (Pass/Fail) · 추가 감사 BLOCK N개
전부 0 이면 "커밋 준비 완료", 하나라도 남으면 "N개 수정 후 재검토"
BLOCK 항목이 있으면 수정 후 에이전트 재위임합니다.
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.
- 5d ago Changed · +7 lines 219613596a6f
- 12d ago First seen · 64 lines · 46 tokens per session scan A aa51bdfa41da
fe-review is a skill published in the GitHub repository sh5623/fe-rail (1 stars, last pushed 7d ago), licensed MIT. It adds 46 tokens to every session and 896 once invoked, about $0.0002 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.
Other skills, from other repositories
spotpatch
A development workflow for React that connects a selected page element to its JSX or TSX source code and gathers context for review.
figma-fetch
Extract frame/component structure and all visible text from a Figma design URL. Use whenever a prompt or a fetched issue contains a figma.com/design or figma.com/file link.
e2e-generate
Create or update end-to-end tests for a user-facing flow that changed, using whatever e2e framework the repo already uses. Use after implementing a user story that alters UI behavior, routing, forms, or API-driven views.
nextjs-patterns
Next.js App Router — Server Components, Actions, streaming, caching. Use when building or migrating Next.js apps.
Component Test Scaffold (Next.js)
Generate React/Next.js component test skeletons (RTL) from specifications.
Next.js App Router Client/Server Boundary
A review rule for Next.js App Router components that checks whether server-side components use browser-only React hooks or browser APIs.