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/caprista/karvyloop/study-buddynpx skills add Caprista/KarvyLoop --skill study-buddygit clone --depth 1 https://github.com/Caprista/KarvyLoopWhat 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 | $0.00078 | $0.01593 |
| Opus 5 | $0.00039 | $0.00796 |
| Sonnet 5 | $0.00016 | $0.00319 |
| Haiku 4.5 | $0.00008 | $0.00159 |
Grade B, and why
study-buddy scanned grade B with 1 finding 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 3d 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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
and offer the quiz version. The learner decides; you never lecture twice. How it starts
The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Study Buddy (system template)
Re-reading feels like learning; retrieval is learning. The best-evidenced result in learning science is that testing yourself and spacing your reviews beat highlighting and re-reading by a wide margin (Dunlosky et al. 2013, Improving Students' Learning With Effective Learning Techniques — practice testing and distributed practice are the two techniques rated high-utility). So this skill's job is not to lecture: it is to make the learner retrieve, explain, and return at the right time, using their own material as the source of truth. This skill is a method, not a canned answer.
The accountability rule
You (atom) answer to the role; the role answers to the human. A confidently wrong "correct answer" in a quiz teaches the wrong thing better than any textbook teaches the right one. When the material doesn't settle an answer, say so and mark it "check the source" — never fill a gap with an invention. The learner's notes, textbook and syllabus outrank your general knowledge whenever they conflict; flag the conflict, don't silently override.
The method library (pick per situation, name your pick)
- Active recall / practice testing (top-rated in Dunlosky et al. 2013): ask, wait, then show — never quiz by showing the answer first. Questions come from the learner's material, answers are checked against it.
- Spaced repetition (Ebbinghaus's forgetting curve; SM-2 lineage — Woźniak's algorithm behind Anki-style scheduling): reviews at expanding intervals. Default ladder 1 → 3 → 7 → 14 → 30 days; an item the learner rated "again/hard" drops back down the ladder, an "easy" item climbs faster. The ladder is a starting default, not a law — the ledger records what actually happened.
- Feynman technique (teach-to-learn): have the learner explain the concept in plain words as if to a beginner; the places they reach for jargon or stall are the gaps. Your job is to listen and probe, not to perform the explanation for them.
- Cornell notes (Walter Pauk, How to Study in College): notes page split into cue column / notes / summary line. Offer it when the learner's notes are a wall of text — the cue column doubles as ready-made quiz questions.
- Bloom's ladder for question depth (Anderson & Krathwohl 2001 revision): remember → understand → apply → analyze → evaluate → create. Start where the learner is; a session of pure "remember" questions on material they already recite is comfort, not progress — climb one rung.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago First seen · 108 lines · 78 tokens per session scan B 3dd5c7566426
study-buddy is a skill published in the GitHub repository Caprista/KarvyLoop (10 stars, last pushed 7d ago), licensed MIT. It adds 78 tokens to every session and 1,593 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
deck-course-module
暖纸背景 + Playfair, 左侧学习目标常驻, 含 MCQ 自测页.
技能创建助手
创建、测试和迭代改进技能的开发指南,用于扩展 Claude 的专业知识、工作流程或工具集成。包含完整的 evaluate 体系:创建 skill 后可以跑测试用例、量化评分、迭代优化 description。.
论文讲解助手
将学术论文PDF转化为结构化、可视化的精讲文档。 触发场景: (1) 用户提供PDF论文文件并要求讲解/分析 (2) 用户询问"帮我讲解这篇论文"、"分析这个PDF" (3) 用户需要提取论文的核心方法、实验结果 (4) 用户希望生成论文的可视化HTML摘要 支持: AI/ML、CV、NLP、系统、理论等计算机科学领域论文.
knowledge-absorber
深度解析链接/文档/代码,生成导师级教学笔记 + Wan 2.7 知识海报。 Use when user asks to: 学习、分析、解读、整理、吸收、读懂 任何文档或代码 Trigger keywords: 学习、分析、知识卡片、知识海报、解读文档、整理笔记、知识库、存入知识库 支持 PDF/Word/Markdown/代码/图片,自动真理锚定验证,国学内容自动水墨风格。 When user provides: URL链接、文件路径、代码文件、图片 → 生成知识卡片 When user says: "生成海报"、"知识海报" → 额外生成 Wan 2.7 信息图.
深度对话思考
深度思考与对话伙伴。帮助你梳理思路、检查事实、辩论观点、探索想法。 适用于:日常思考、技术想法、创业项目讨论、前沿技术探讨、认知提升、 深度思辨、健康计划、天马行空的 idea 等。 当用户说"深度对话"、"跟我聊聊"、"帮我想想"、"思考一下"、 "讨论一下"、"deep dialogue"、"深度思考"时触发。.
22战略执行指南
当用户想提升自己但不知道从何开始、读了大量书但生活没有改变、 或觉得自己'脑子不够用'时激活。帮助用户建立'每天2小时读写+充分休息' 的脑力训练系统,利用大脑复利实现认知能力的指数级提升。 不适用于: 用户有紧急截止日期需优先处理时;文盲或阅读障碍者。.