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 agentmods add skills/ggombee/code-forge/figma-to-codenpx skills add ggombee/code-forge --skill figma-to-codegit clone --depth 1 https://github.com/ggombee/code-forgeWhat 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 | $0.00020 | $0.00878 |
| Opus 5 | $0.00010 | $0.00439 |
| Sonnet 5 | $0.00004 | $0.00176 |
| Haiku 4.5 | $0.00002 | $0.00088 |
Grade A, and why
figma-to-code 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Figma to Code Skill
Figma 디자인을 Emotion styled 기반 코드로 변환
트리거 조건
| 트리거 | 반응 |
|---|---|
| Figma URL 제공 | 즉시 실행 |
| "Figma 변환", "디자인 구현" | 스킬 활성화 |
/start 커맨드 내 Figma 분석 |
자동 연계 |
워크플로우
Phase 1: 디자인 추출
// Figma MCP 도구로 디자인 컨텍스트 추출
mcp__figma__get_design_context({
fileKey: '{file_key}',
nodeId: '{node_id}',
clientLanguages: 'typescript',
clientFrameworks: 'react',
});
추출 항목:
- 레이아웃 구조 (flex, gap, padding)
- 색상 값 (hex → semantic color 매핑)
- 타이포그래피 (font-size, weight, line-height)
- 컴포넌트 계층 구조
Phase 2: 컴포넌트 매핑
파일 구조:
{컴포넌트}/
├── index.tsx # 컴포넌트 로직
└── styled.ts # Emotion styled 정의
styled.ts 템플릿:
import styled from '@emotion/styled';
export const Container = styled.div`
display: flex;
flex-direction: column;
gap: 16px;
padding: 24px;
`;
// 상태별 스타일 (transient props: $접두사)
export const FilterButton = styled.button<{ $isSelected: boolean }>`
padding: 8px 16px;
`;
index.tsx 템플릿:
import { FC } from 'react';
import { Container, Title, FilterButton } from './styled';
interface Props {
// props 정의
}
export const Component: FC<Props> = ({ ...props }) => {
return (
<Container>
{/* styled 컴포넌트 직접 사용 */}
</Container>
);
};
Phase 3: Agent Teams 병렬 구현 (Claude Max)
여러 컴포넌트를 동시에 생성할 때:
TeamCreate({ team_name: 'figma-team', description: 'Figma to Code 변환' });
Task(subagent_type='implementor', team_name='figma-team', name='component-1', model='sonnet',
prompt='FilterButton 컴포넌트 구현');
Task(subagent_type='implementor', team_name='figma-team', name='component-2', model='sonnet',
prompt='DateControls 컴포넌트 구현');
// 완료 후 → shutdown_request → TeamDelete
Agent Teams 미가용 시 Task 병렬 호출로 폴백
Phase 4: 검증
Pixel-perfect 체크리스트:
- 레이아웃 (flex-direction, gap, padding) 일치
- 색상 (semantic color 매핑) 일치
- 폰트 (size, weight, line-height) 일치
- 상태별 스타일 (hover, active, disabled) 일치
- 반응형 동작 (있는 경우)
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.
- yesterday First seen · 144 lines · 20 tokens per session scan A 9273db466ad4
figma-to-code is a skill published in the GitHub repository ggombee/code-forge (13 stars, last pushed 2mo ago), licensed MIT. It adds 20 tokens to every session and 878 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
software-implementation
Write the minimal production code needed to make failing TDD tests pass (Green phase of red-green-refactor). Use when failing tests exist and production code must be written or modified to satisfy them. Reads and interprets failing test output, discovers codebase conventions (module layout, naming, import patterns…
plan
Plan mode: write markdown plan to .mateclaw/plans/, no exec.
pypto-pitfalls
:标量任意位置。- /:tensor 必须在左。函数调用:第一参数必须 Tensor。.
tilelang-cuda-memory
TileLang CUDA 内存访问优化策略,包括 T.allocshared/fragment 使用、数据布局优化、合并访存和 Bank Conflict 避免技巧。适用于内存带宽受限、需要优化数据搬运效率的 TileLang 内核性能优化场景.
tilelang-cuda-synchronization
TileLang CUDA 同步规范,包括 T.syncthreads() 使用规则、线程安全最佳实践和死锁预防策略。适用于编写涉及共享内存访问、多线程协作、或需要避免同步死锁的 TileLang 内核代码生成场景.
pypto-loop-view
最常见的首次生成错误。所有 pypto.view 的 shape 参数必须是编译期常量(字面量或闭包变量)。.