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 renky1025/agent-skills --skill graspgit clone --depth 1 https://github.com/renky1025/agent-skillsWrote 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/renky1025/agent-skills/grasp)<a href="https://agentmods.dev/skills/renky1025/agent-skills/grasp"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/grasp/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/renky1025/agent-skills/grasp"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/grasp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00149 | $0.02121 |
| Opus 5.5 | $0.00060 | $0.00848 |
| Sonnet 5.5 | $0.00030 | $0.00424 |
| Haiku 4.5 | $0.00015 | $0.00212 |
Grade A, and why
grasp 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 1mo 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Grasp — 十维认知框架 × 加速学习协议
Purpose
把任意主题压缩成可复述、可复习、可继续深入的认知资产。以十维认知框架(名称/类别/定义/特征/组成/作用/条件/历程/前景/风险)为内容骨架,嵌入 Feynman 讲解、第一性原理拆解、主动回忆和间隔重复四种学习技术,产出知识地图 + 维度诊断矩阵 + 分阶段路线图。
When to Use
- 用户要系统吃透一个陌生主题或领域,而不是要一个孤立事实。
- 用户已有基础,想检测盲区或规划进阶路线。
- 用户要一份可复习的知识地图或学习路线图。
- 触发词:
/grasp <topic>、grasp、速通、拆解、吃透、掌握、学习路线图、study guide、learn fast、master [topic]、Feynman technique、active recall。 - 需要路由时:
--phase=N表示运行到第 N 阶段后停止。
When NOT to Use
- 用户只需要一个快速事实 -> 直接回答,不启动协议。
- 纯操作技能 / 肌肉记忆(需要练习而非认知框架)-> 交给练习设计,本技能不适用。
- 用户明确只要某一维度的信息 -> 直接回答该维度,不跑全协议。
- 用户要"把一件事给外行讲明白"的图文教学页 -> 交给
teach-eli5。 - 用户要设计正式课程、教学大纲或教案 -> 交给
curriculum-design。
Workflow
七阶段顺序执行。每阶段结束时输出签退问题,用户确认后进入下一阶段;每阶段的完整引导问题与产出格式见 references/ten-dimensions.md。
- Identity(维度 1-4) — 用 Feynman 方式讲解名称、类别、定义、特征(避免术语、使用类比),随后要求用户 teach-back,标出准确 / 遗漏 / 混淆点。
- Anatomy(维度 5-6) — 拆解组成与结构(列组件、绘关系、分核心与辅助、展示因果链),分析核心功能与作用机制,用第一性原理验证(去掉某核心组件会怎样)。
- Ecosystem(维度 7-8) — 用"生存条件"框架分析必要条件 / 理想条件 / 失效条件 / 互操作条件;构建含至少 3 个转折点的时间线,做驱动力与遗产分析。
- Horizon(维度 9-10) — 研判当前阶段(萌芽/增长/成熟/衰退)与 2-3 条发展方向,按内在局限 / 实践瓶颈 / 社会伦理 / 失败模式四类审查风险,做一次批判性反问(做空论证)。
- Recall(全 10 维) — 脱离资料,逐维提问(难度递增),生成维度诊断矩阵并路由补救。
- Map(全 10 维) — 将十维成果压缩为一页知识地图,并制定间隔重复计划。
- Path(全 10 维) — 评估当前水平,输出分阶段路线图 + 资源推荐 + 一条可立即执行的下一步。
Decision Rules
- 用户给
--phase=N-> 只运行到第 N 阶段并停止(--phase=5直接跳到全维度诊断,--phase=6只出知识地图,--phase=7只出路线图)。 - 用户未指定 -> 从 Phase 1 完整走完 1-7。
- 用户目标含糊或无明显基础 -> 先跑 Phase 1,不要跳过身份层。
- Phase 5 诊断出问题 -> 按类型路由:
| 问题类型 | 路由 |
|---|---|
| 概念混淆、关系缺失 | 回到对应 Phase 重新讲解 |
| 信息不足 | 记入 Phase 7 学习路线图 |
| 记忆消退 | 纳入 Phase 6 间隔重复 |
| 推理错误 | 回到 Phase 2 用第一性原理重新推导 |
- 需要某阶段的引导问题或提问模板 -> 读
references/ten-dimensions.md。 - 需要知识地图 / 路线图的逐字模板 -> 读
references/deliverable-templates.md。 - 用户追问为什么这样学、或需要复习排期 -> 读
references/learning-techniques.md。
Constraints
- 十维覆盖是底线:每个 Phase 明确标注覆盖维度,不得遗漏。
- 方法服务于维度:Feynman、第一性原理是手段,不是独立阶段。
- Phase 1-4 必须按认知深度递进,不得平铺。
- 不得编造事实、来源或数字;信息不足时标注"信息不足"并记入路线图,而不是填充内容。
- 讲授语言跟随用户;术语首次出现必须就地解释。
What ships with it
3 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.
- 1mo ago Changed · -236 lines · +149 tokens per session dd62084f99fc
- 1mo ago First seen · 345 lines · 0 tokens per session scan A 89eab05d7a3d
grasp is a skill published in the GitHub repository renky1025/agent-skills (11 stars, last pushed 15d ago), licensed MIT. It adds 149 tokens to every session and 2,121 once invoked, about $0.0006 per session on Opus 5.5. 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
mindos
MindOS: local knowledge assistant & shared KB. Keeps decisions, notes, SOPs, debugging lessons, research findings, preferences across sessions/agents. Core: save notes, search KB, organize files, run workflows, review, append CSV, hand off context, distill lessons. NOT for app source or paths outside KB. Triggers…
hidden-talents
Interview you in depth to surface overlooked or suppressed talents and write your Personal Talent Manual.
socratic
Diagnose a confusion Socratically, one question at a time, until the question actually worth answering surfaces — then answer that one.
two-layer-explain
Explain a concept twice — for a newcomer and for an expert — then bridge the two layers, flag the likely misunderstandings, and set three check questions.
deconstruct
Reverse-engineer why a finished piece of work succeeds and extract what transfers to your own — rules, a checklist, and a first exercise.
newbie-guide
A Chinese-language guide for beginners who have an unclear goal or do not know how to break a task into steps.