using-karpathy-wiki
Auto-loaded by the SessionStart hook. Defines when to capture and routes the main agent or detached ingester to the appropriate on-demand skill.
供给资产档案
研究与知识工作
Deep research, source comparison, literature review, RAG, knowledge search, and reports.
场景
研究 Agent
I need my agent to research a topic, compare sources, and produce a concise report.
适配 Agent
Claude Code + OpenAI Agents + CLI
适用于 Codex、Claude Code、Cursor、CLI 或自定义 Agent。
安装
就绪
npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wiki
维护状态
新鲜
距上次推送 3 天
风险
需审查
Permission surface may require sandboxing
GitHub 质量
101
66/100 质量 · 76/100 信任
覆盖标签
审查说明
Permission surface may require sandboxing · Financial research output is not financial advice; require human review before any live investment decision
Agent 采用评分卡
一眼查看信任、审计与安装准备度
这些分数综合公开仓库元数据、OpenAgentSkill 审查信号、维护新鲜度与安装准备度。它用于候选筛选,不替代人工审查。
质量
有潜力有用的候选项,但采用前应与替代方案比较。
信任
仅限沙盒有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。
审计
需审查对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。
OpenAgentSkill 信任评分 v5
安装前需人工审查
仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。
Stars
101 个 GitHub Stars
仓库活跃度
101 个 Star,24 个 Fork
维护状态
距上次推送 3 天
许可证
MIT
安装
npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wiki
安装安全性
标准软件包或运行时安装路径
权限范围
shell or command execution, filesystem or document access
Agent 结果
暂未有 Agent 结果数据
文档
README/SKILL.md 上下文充分
风险摘要
生产前审查
- Financial research output is not financial advice; require human review before any live investment decision.
- Quality score needs review
- Permission surface needs review: shell or command execution, filesystem or document access
- Stars/forks activity: 101 stars, 24 forks; issue activity unavailable in current metadata
安装准备度
安装路径可用
- 安装路径可用
- 仓库证据可用
- 已声明许可证
- 暂无 Agent 验证结果证据
Agent 可读元数据
这个 Skill 的机器可读决策数据。
使用此区块或内嵌 JSON 判断 Agent 是否应安装该 Skill、选择替代方案,或先请求人工审查。
适用任务
- 研究 Agent 工作流
- Claude Code 团队
- builders willing to evaluate younger projects
- 检索来源
适用 Agent
安装决策
- 命令
- npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wiki
- 策略
- 审查
- 人工审查
- 是
信任与风险
- 信任
- 68/100
- 审计
- 79/100
- 风险级别
- 需审查
结果闭环
- 端点
- /api/agent/outcome
- 事件 ID
- resolve
- 结果
- 5
不适用场景
- 需要厂商支持 SLA 的团队
- 没有内部安全审查的高合规环境
- 暂未有 OpenAgentSkill 使用反馈数据
- 高风险权限提示:Shell 或命令执行
- Permission surface may require sandboxing
Agent 安全 v2
47/100 · 避免自动安装
Sparse or mixed signals. Useful for discovery, but not for autonomous installation.
Test manually in an isolated workspace and compare against safer alternatives.
高
Shell 或命令执行
Skill 元数据引用了终端、CLI、Shell、子进程或命令执行工作流。
中
网络访问
Skill 可能访问远程页面、API、仓库或外部服务。
中
文件系统访问
Skill 可能读取或写入项目文件、文档、生成产物或本地工作区状态。
中
数据库访问
Skill 可能检查 Schema、查询数据库或处理持久化存储。
- 高风险权限提示:Shell 或命令执行
- Permission surface may require sandboxing
安装目标
在你的 Agent 工作流中安装此 Skill
通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。
OpenAgentSkill CLI
Resolve policy, run the source installer safely, and report a verified install receipt.
$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install toolboxmd-using-karpathy-wikiAgent 解析计划
让 Agent 在安装前验证匹配度。
Resolve API 返回首选 Skill、替代方案、安全策略、审计说明、安装目标和可直接执行的提示词,无需抓取此页面。
打开 JSON
/api/agent/resolve?task=Use%20using-karpathy-wiki%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Resolve 文本
/api/agent/resolve?task=Use%20using-karpathy-wiki%20for%20an%20agent%20workflow&agent=codex&max_risk=medium&format=text
安装交接
/api/skills/toolboxmd-using-karpathy-wiki/install
Agent 应检查
- 从 Resolve API 检查任务匹配与替代方案。
- 检查审计评分、信任评分和安全策略警告。
- 检查 Codex、Claude Code、Cursor 或 CLI 的安装目标兼容性。
复制提示词
Task: Use using-karpathy-wiki in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20using-karpathy-wiki%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/toolboxmd-using-karpathy-wiki/install
Install command: npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wiki
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.Agent 交接
把安装路径交给 Agent,而不是再给一个目录页。
通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。
安装交接
/api/skills/toolboxmd-using-karpathy-wiki/install
LLM 文本格式
/api/skills/toolboxmd-using-karpathy-wiki/install?format=text
寻找替代方案
/api/skills/search?q=using-karpathy-wiki&limit=3
Agent 提示词
Use using-karpathy-wiki for this task. Review https://www.openagentskill.com/api/skills/toolboxmd-using-karpathy-wiki/install, then install with: npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wikiRegistry 元数据
用于自动选择 Skill 的 Agent 可读档案。
本页通过 Registry API 提供相同的决策、信任、审计、场景和安装信号,让 Agent 无需抓取界面即可排序。
Agent 决策面板
Fallback candidate for Research agents
先用此 Skill 做原型验证,并保留备选方案。
栈中角色
备选候选
主要匹配
研究 Agent
信任标签
先做原型验证
安装路径
命令已就绪
适用场景
- 研究 Agent 工作流
- Claude Code 团队
- builders willing to evaluate younger projects
证据
- 仓库近期活跃
- 已提供安装命令或 GitHub 仓库
- 66/100 质量档案
先审查
- 暂未有 OpenAgentSkill 使用反馈数据
实施路径
- 1在沙盒 Agent 中安装它,并端到端完成一次研究 Agent任务。
- 2Compare output quality, latency, and failure behavior against at least one alternative.
- 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.
信任档案
仅限沙盒
有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。
GitHub 采用度
信息101 个 GitHub Stars
Star/Fork 活跃度
检查101 个 Star,24 个 Fork; 当前元数据中没有议题活跃度信息
近期维护
通过距上次推送 3 天
许可证清晰度
通过MIT
积极信号
- AI 审查已通过
- 安装路径可用
- 仓库证据可用
- 近期维护的仓库
- 安装命令未发现明显高风险模式
- 结果闭环已就绪,但需要首次真实 Agent 运行
安装前审查
- Financial research output is not financial advice; require human review before any live investment decision.
- Quality score needs review
- Permission surface needs review: shell or command execution, filesystem or document access
- Stars/forks activity: 101 stars, 24 forks; issue activity unavailable in current metadata
- Permission surface: shell or command execution, filesystem or document access
- 暂未有真实 Agent 结果报告
- 无人值守安装前需要人工审查
建议操作
仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。
质量档案
有潜力 适用于 Agent 工作流的候选
有用的候选项,但采用前应与替代方案比较。
工作流匹配
在这些场景使用此 Skill
Investigate faster
Research agents
I need my agent to research a topic, compare sources, and produce a concise report.
Build and ship code
Coding agents
I need a coding agent that can understand a repository, edit code, and review pull requests.
Search private knowledge
RAG and knowledge
I need my agent to build a RAG workflow over documents and retrieve reliable context.
工作流匹配
加入完整工作流
Find, compare, and synthesize
Research report agent
A workflow for agents that gather sources, compare claims, summarize long material, and draft useful research briefs.
Inspect, patch, and verify code
Coding review agent
A workflow for software agents that inspect repositories, review pull requests, generate tests, and turn findings into shippable patches.
Ingest, retrieve, and cite
RAG knowledge base
A workflow for document-heavy agents that ingest files, create searchable knowledge, retrieve relevant context, and answer with grounded sources.
替代方案短名单
安装前对比
可能适合该任务的相近 Skill。
UI-TARS Desktop
Run multimodal agents that operate desktop interfaces
MoneyPrinterTurbo
利用AI大模型,一键生成高清短视频 Generate short videos with one click using AI LLM.
Cua
Open-source infrastructure for Computer-Use Agents. Sandboxes, SDKs, and benchmarks to train and evaluate AI agents that can control full desktops (macOS, Linux, Windows).
概览
--- name: using-karpathy-wiki description: | Auto-loaded by the SessionStart hook. Defines when to capture and routes the main agent or detached ingester to the appropriate on-demand skill. ---
<SUBAGENT-STOP> If you were dispatched as a subagent for a task unrelated to wiki capture or ingest, skip this skill. The wiki rules apply to the main agent in the conversation; subagents doing isolated work (Explore, code-reviewer, codex:rescue, etc.) do not need to capture. </SUBAGENT-STOP>
# karpathy-wiki — loader (read once per session, applies for the whole session)
You are operating in a project that uses the karpathy-wiki plugin. Durable knowledge surfaced in this conversation must be captured to a wiki so future sessions can find it.
## Announce when you act
When you write a capture, run an ingest, or answer from the wiki, prefix your reply with one line:
> **Using the karpathy-wiki skill to [capture this / ingest pending captures / answer from wiki].**
This is the ONLY wiki-mechanics text the user sees. Do not narrate orientation, capture authoring, dispatch mechanics, or state-machine progress. Do all the wiki work silently after the announce line.
## Instruction priority
If the user's `CLAUDE.md` / `AGENTS.md` or an explicit user instruction conflicts with the rules below, follow the user. The wiki rules override the default system prompt where they conflict, but **never** override the user's direct instructions.
## Iron laws
``` NO WIKI WRITE IN THE FOREGROUND NO PAGE EDIT WITHOUT READING THE PAGE FIRST NO SKIPPING A CAPTURE BECAUSE "IT DOESN'T LOOK WIKI-SHAPED" NO ANSWERING ANY USER QUESTION WITHOUT ORIENTING FIRST ```
Captures go to `<wiki>/.wiki-pending/<timestamp>-<slug>.md`. The bounded dispatcher assigns each capture to the configured provider/model profile and ingests it in the background.
## Orient before answering — the read protocol
For ANY user message that is asking a QUESTION (vs requesting a code edit, file operation, or pure command), you must orient on the wiki BEFORE drafting the answer. Pre-classifying a question as "general knowledge" or "not wiki-relevant" is the forbidden rationalization Iron Rule 4 was written to forbid.
**Cost framing.** Orientation is a once-per-session investment:
- First wiki-eligible question of the session: read `<wiki>/schema.md` and the relevant `<wiki>/<category>/_index.md` (or root `<wiki>/index.md` for cross-category questions). Two file reads, ~10-30 KB total. - Subsequent questions in the same session: the schema and index are already in your working memory; the marginal check is a candidate-count scan against the (already-read) `_index.md`. Near-zero cost.
Load `karpathy-wiki-read/SKILL.md` for the deterministic 6-step ladder (orient → count candidates → inline-read OR subagent OR web-search → cite). Do NOT skip the load because "this question seems trivial" — the iron rule is unconditional.
**Resist-table for the read protocol** (capture-side resist-table is below, after TRIGGER):
| Rationalization | Reality | |---|---| | "This is general knowledge / I know this from training" | Pre-classifying questions as wiki-irrelevant is the forbidden rationalization. The wiki's scope is whatever has been captured. You don't know without checking. Orient. | | "This question is trivial; reading two files is overkill" | Orientation is once per session. After the first orient, the marginal cost on the next question is near-zero (a re-scan of the already-read `_index.md`). The "trivial question" carve-out is the same drift that dropped the read protocol in v2.4. | | "Pure syntax question, the wiki won't cover this" | You cannot know without checking. If `_index.md` has zero matches, the read skill's Step F handles it (web search + capture the gap). The orient itself is what proves there are no candidates. | | "User asked me to fix a bug, not answer a question" | Code-edit requests don't trigger the read protocol; question-shaped requests do. If unsure (e.g. "why isn't this working" — could be a question OR a debug request), orient. | | "I already oriented earlier this session, I'll skip this time" | Correct! Re-use the schema and index you already have. Run only the candidate-count check on the new question's terms. The protocol's deterministic part (Step B) is the only thing that re-runs per question. |
## TRIGGER — when to capture
Write a capture when ANY of these fire (immediate-capture triggers):
- A research subagent or research agent completes or returns a file with durable findings. - New factual information is found (web search, doc fetch, external fact surfaced in conversation). - Session resolves a confusion — something neither of you knew before. - A gotcha, quirk, or non-obvious behavior in a tool is discovered. - A pattern is validated (approaches compared, one picked, with reasons). - An architectural decision is made with explicit rationale. - User pastes a URL or document for study. - `<wiki>/inbox/` or `<wiki>/raw/` has unprocessed files. - User says "add to wiki" / "remember this" / "wiki it" / "save this". - Two claims contradict each other.
Also TRIGGER (orientation + citation): the user asks "what do we know about X" / "how do we handle Y" / "what did we decide about Z" / "have we seen this before" — or any question the wiki might cover.
## SKIP — when not to capture
- Routine file edits, syntax lookups, one-off debugging with trivial root causes. - Time-sensitive data that must be fetched fresh. - Questions clearly outside any wiki's scope.
Do NOT skip based on tone or shape. "This looks like casual chat", "there's no code here", "this isn't a wiki context" are forbidden rationalizations. If new factual information appeared, capture. Tone is not the trigger.
**Resist-table for capture triggers.** When you're about to skip a capture, check these red flags. Skipped captures are invisible — the user cannot observe them, so the discipline is yours alone.
| Rationalization | Reality | |---|---| | "The user will remember this" | The user will not remember. That's the whole point. | | "It's too trivial for the wiki" | If a TRIGGER fires, capture. Lint and the ingester filter noise later. | | "I'll capture it later" | Later means never. Capture now — invocation is milliseconds. | | "I'm in the middle of another task" | Capture is non-blocking; `bin/wiki capture` writes a file and returns. The ingester runs detached. | | "The user didn't ask me to save this" | Triggers fire automatically. No explicit user request is required — that's the loader's contract. | | "I don't have a memory tool available" | This skill IS the memory tool. The loader's presence is the trigger. | | "The file is already in a good place" | Filing ≠ capturing. Location is not organization. The capture extracts the durable concept from the file's context; the file alone doesn't. |
## What to do when a trigger fires
For chat-driven captures (the conversation IS the source, or you have a file alongside the conversation): load `karpathy-wiki-capture/SKILL.md` and follow it. The capture is written via `bin/wiki capture` (a subcommand of the existing `bin/wiki` CLI).
For research subagent reports (a file is the source): the agent's job is one command, NOT a hand-written capture body:
```bash mv <subagent-report-path> <wiki>/inbox/<basename> wiki ingest-now <wiki> # or wait for next SessionStart ```
For ANY user question (including "what does the wiki know about X" questions, but also any other question per Iron Rule 4): load `karpathy-wiki-read/SKILL.md` and run the deterministic 6-step ladder. Do NOT load `karpathy-wiki-ingest/SKILL.md` — that's for the detached runtime ingester only and contains write-side machinery the main agent never needs.
## Mode change
If the user asks to change the wiki mode for this directory (phrases like "use project wiki here," "main only," "switch to both," "fork to main"), run:
```bash wiki use project|main|both ```
Confirm the change in one line; do not over-narrate.
## What's NOT in this loader
The full operational details — capture format, body-size floors, dispatch mechanics, ingester orientation, page format, manifest protocol, commit conventions — live in the on-demand skills:
- `skills/karpathy-wiki-capture/SKILL.md` — load when you are about to write a capture. - `skills/karpathy-wiki-read/SKILL.md` — load when you are about to answer a user question (Iron Rule 4). - `skills/karpathy-wiki-ingest/SKILL.md` — loaded by the detached ingester via its provider prompt; the main agent never reads this.
If you do not know what to do at a particular step, load the on-demand skill — do not invent.
技术详情
- 版本
- 1.0.0
- 许可证
- MIT
- 最近更新
- 2026年8月20日
- 发布时间
- 2026年8月20日
决策摘要
备选候选
仓库近期活跃
Agent 验证证据
Agent 验证证据
来自解析、审查、安装和一次小范围运行后的结果报告。
- 成功率
- —
- 近期失败
- —
- 结果
- 0
- 输出质量
- —
- 失败
- 0
- 不相关
- 0
- 安装次数
- 0
- 风险拦截
- 0
- 需要配置
- 0
- 生产环境
- 0
暂时没有 Agent 结果数据。首次 Agent 执行可以通过 /api/agent/outcome 报告成功、需要设置、风险拦截、失败或不相关。
增长闭环
分享工具包
为 using-karpathy-wiki 准备的场景化草稿,可手动发布到 X。
using-karpathy-wiki: Auto-loaded by the SessionStart hook. Defines when to capture and routes the main agent or de... 101 stars https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki?ref=x
可选:带安装命令的回复
Listing + install path for using-karpathy-wiki: https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki?ref=x Install: npx skills add toolboxmd/karpathy-wiki --skill using-karpathy-wiki
收录来源
Registry 收录
此列表来自公开来源,维护者认领获批前不会标记为官方。
- 创作者
- toolboxmd
- 收录方
- OpenAgentSkill 社区索引
归属链接指向公开仓库或创作者主页。创作者可认领列表以更新所有权信号。
认领此 Skill所有者认领
认领此 Skill 页面
这条 Registry 收录 列表归属于 toolboxmd,但尚未标记为官方。认领后可增加已验证所有者信号,使后续发布、安装和审计更新更值得信赖。
创作者外链工具包
将证据徽章加入你的 README
在开发者评估仓库的位置展示规范页面、当前信任与审计信号,以及真实的 Agent 验证证据。
[](https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki)
[](https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki)
[](https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki/audit)
[](https://www.openagentskill.com/skills/toolboxmd-using-karpathy-wiki)作者
toolboxmd
@toolboxmd
健康信号
- GitHub Stars
- 101
- 质量评分
- 37/100
- 最近 GitHub 推送
- 2026年8月20日
- 框架提示
- 未知
- OpenAgentSkill 浏览量
- 0
- 复制安装命令
- 0
- 跳转点击
- 0
社区信号
告诉我们这个 Skill 是否对你的 Agent 工作流有帮助。汇总反馈会持续改善排序。
信任与安全
仅限沙盒
- GitHub 采用度101 个 GitHub Stars信息
- Star/Fork 活跃度101 个 Star,24 个 Fork; 当前元数据中没有议题活跃度信息检查
- 近期维护距上次推送 3 天通过
- 许可证清晰度MIT通过
- README/SKILL.md 完整度元数据包含足够的用法与工作流上下文通过
- 依赖与运行时风险command execution surface, network or browser surface信息
相关 Skill
UI-TARS Desktop
Run multimodal agents that operate desktop interfaces
37.0K StarsMoneyPrinterTurbo
利用AI大模型,一键生成高清短视频 Generate short videos with one click using AI LLM.
88.5K StarsCua
Open-source infrastructure for Computer-Use Agents. Sandboxes, SDKs, and benchmarks to train and evaluate AI agents that can control full desktops (macOS, Linux, Windows).
21.4K Stars