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/itlearning/study-ios/swift-studynpx skills add ITlearning/study-ios --skill swift-studygit clone --depth 1 https://github.com/ITlearning/study-iosWhat 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 | $0.00027 | $0.05646 |
| Opus 5 | $0.00014 | $0.02823 |
| Sonnet 5 | $0.00005 | $0.01129 |
| Haiku 4.5 | $0.00003 | $0.00565 |
Grade A, and why
swift-study 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 510 lines — stays where its author put it; the contents beside it link to each section on GitHub.
swift-study - Interactive Swift/iOS Learning
Swift/iOS tutor with two learning styles: (두 가지 학습 방식을 지원하는 Swift/iOS 튜터)
- Classic: explanation-first flow for quick understanding
- Mastery: predict-first Core Loop with gate checks for deeper training
Phase 0: Setup
Terminology Policy (초심자 친화 용어 규칙)
Use Korean-first labels for learning steps. Keep English terms only as secondary references in parentheses. (학습 단계 이름은 한국어를 우선 사용하고, 영어 용어는 괄호로 보조 표기한다.)
Required style: (표기 규칙)
예측하기 (PREDICT)해설하기 (REVEAL)한 줄 정리 (RESTATE)학습 관문 (Gate)통과 / 보완 필요instead of rawPASS / FAIL
When a technical term appears for the first time, add a one-line plain explanation immediately. (기술 용어가 처음 나오면 바로 다음 줄에 쉬운 말 한 줄 설명을 붙인다.)
Example: (예시)
격리 경계 (isolation boundary): "이 선 밖에서는 actor 내부 상태를 직접 건드릴 수 없다."경합 상태 (data race): "여러 작업이 같은 값을 동시에 바꾸려고 해서 결과가 꼬이는 문제."
Language Selection
Ask the user to choose a language at the start using a selection flow, not free-form text input. (스킬 시작 시 자유 입력이 아니라 선택형 방식으로 언어를 받는다)
- 한국어 - 한국어로 학습합니다
- English - Learn in English
Use platform-specific selection behavior: (플랫폼별 선택 방식)
- Codex CLI: use native option selection (
request_user_inputwith 2 options) when available. If unavailable, show numbered choices and ask for1or2. - Claude Code: use native option selection UI when available. If unavailable, show numbered choices and ask for
1or2. - Gemini CLI: use native option selection UI when available. If unavailable, show numbered choices and ask for
1or2.
Choice mapping: 1 → 한국어, 2 → English
After a valid choice, use the selected language for all communication. Code and Swift keywords stay in English. (선택한 언어로 이후 모든 소통을 진행한다. 코드와 Swift 키워드는 영어 그대로 유지한다.)
Study Mode Selection
Ask the user to choose a study mode using the same selection pattern: (같은 선택 방식으로 학습 모드를 선택하게 한다)
- Classic (설명 중심, 추천)
- 처음 배우는 주제를 빠르게 이해할 때 적합
- 튜터가 먼저 설명하고, 질문으로 이해도를 확인
- 부담이 적고 진입이 쉬운 흐름
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.
- 2d ago First seen · 510 lines · 27 tokens per session scan A 1203facdb31e
swift-study is a skill published in the GitHub repository ITlearning/study-ios (19 stars, last pushed 6mo ago), licensed MIT. It adds 27 tokens to every session and 5,646 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
discussion
Skill "discussion" from naver/design-to-ui, covering discussion — 사용 경험을 discussions로 축적, 카테고리 (등록 대상), steps, step 1: 의도·코멘트 파악 and step 2: 정보 수집 (이 스킬의 핵심 — 3개 출처, 날조 금지).
canvas-homework
Check Canvas for outstanding assignments and either surface the assignment, or draft a response in Max's voice. Use whenever Max asks what assignments / homework / coursework he still has to do, asks to pull an assignment's instructions or files from Canvas, or asks to start/draft an assignment. Orchestrates the…
project-memory
Maintain persistent project memory in MEMORY.md via MCP tools. SCOPE: codebase/architecture/conventions/gotchas + Recent Sessions of code work, in /MEMORY.md. DISTINCT from Claude Code's auto memory (/.claude/projects/ /memory/, holds user profile/preferences/feedback). Code-about → here; user-about → auto memory.…
chaos-engineer
Use when designing chaos experiments, implementing failure injection frameworks, or conducting game day exercises. Invoke for chaos experiments, resilience testing, blast radius control, game days, antifragile systems.
kotlin-specialist
Use when building Kotlin applications requiring coroutines, multiplatform development, or Android with Compose. Invoke for Flow API, KMP projects, Ktor servers, DSL design, sealed classes.
react-native-expert
Use when building cross-platform mobile applications with React Native or Expo. Invoke for navigation patterns, platform-specific code, native modules, FlatList optimization.