Token导航 LogoToken导航

Agent Skill · 研究检索

kb-learn(知识库学习)

kb-learn 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

kb-learn 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为执行命令;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 68,884。

总安装 222周安装 9GitHub Stars 3下载 70核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:kb-learn(知识库学习)
来源仓库:https://github.com/farzadshbfn/knowledge-skills
仓库路径:skills/kb-learn
安装命令:
npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

kb-learn 用于查找、检索和筛选相关信息,适合快速定位候选结果。适用于关键词搜索、任务场景匹配或来源线索筛选等研究检索场景。通过 npx skills add 命令从 GitHub 仓库安装并使用。安装前需确认权限范围、维护状态及是否触发联网或文件操作。建议结合原始 README 核验具体用法和功能边界。

来源仓库内路径:skills/kb-learn

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要执行命令,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

kb-learn 怎么安装?

在终端执行 npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

kb-learn 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

kb-learn 需要什么权限?

权限级别是执行命令。安装流程涉及命令执行,可能通过 npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

kb-learn 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

kb-learn 用的人多吗?

公开数据是累计安装 222 次,近一周安装 9 次,下载 70 次,GitHub 3 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

kb-learn 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
kb-learn

知识库学习 · 本页

2223执行命令未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

/kb-learn

Contents

1. Configuration

On first invocation, check for config at .claude/knowledge-base/config.json.

No config: Tell user "Run /kb-bootstrap to set up your knowledge base." and stop.

Config exists: Read kb_roots. Single entry → use it. Multiple → defer to §4 step 0.

2. Routing

Parse $0 to determine workflow:

2a. Agents

Specialized agents isolate high-volume work and enable parallelism. SKILL.md is the orchestrator — spawns agents and chains results. Agents never spawn other agents.

AgentModelPurposeUsed by
scouterhaikuKB discovery via /kb-find (supports --challenge)all workflows
searchersonnetWeb research — search, fetch, structure findingstopic, article
challengersonnetAdversarial web research — finds counter-evidencetopic
assessoropusEvaluate claims against KB + web evidencearticle, topic

Orchestration Pipeline

1. Spawn scouter(s) + searcher IN PARALLEL
   ├── scouter (normal) → supporting evidence
   ├── scouter (challenge) → counter-evidence
   └── searcher → web findings
2. Spawn challenger (with claims from step 1) → counter-evidence from web
3. Spawn assessor (with all evidence incl. challenger) → verdicts + KB impact
4. Apply changes

Not all steps required every time:

  • Article: scouter(s) + searcher → assessor (no challenger)
  • Topic: full pipeline including challenger
  • Fix: single scouter, no web search

Use agents when output is high-volume, work is parallelizable, or task benefits from model specialization. Use inline when KB is small, single simple claim, or latency matters.

3. Shared Procedure: KB Discovery

All KB discovery goes through the scouter agent. Never use /kb-find directly.

How to spawn scouters

  1. Read the kb-find SKILL.md at ${CLAUDE_SKILL_DIR}/../kb-find/SKILL.md
  2. Build prompt: scouter body text (after frontmatter) + separator + full kb-find SKILL.md content
  3. Spawn Agent with model: haiku, background: true

Why: Skill-internal agents don't get automatic skill injection — the orchestrator must explicitly inject kb-find content.

Spawning patterns

  • Bare lookup (topic, fix): single scouter, normal mode
  • Claim evaluation (article, deep assessment): two scouters in parallel (normal + challenge)

Per-workflow queries

  • article: search by claims and key terms
  • topic: search by topic name
  • fix: search by described error (with scrutiny)

Empty KB: Skip loading, note KB is empty.

3a. Shared Procedure: Concept Evolution and Legacy Patterns

After loading KB, check for concept evolution — outdated terminology, framing, or versioning. Follow reference/evolution-workflow.md for detection, callouts, old patterns, and legacy notes.

4. Shared Procedure: Update KB

  1. KB routing: Multiple KBs → infer target from context; if ambiguous, ask user.

- Read-only guard: If target KB has [read-only] badge (global KB), do NOT write to it. Instead, create a suggestion file in ~/.claude/knowledge-base/suggestions/ using suggestion.py (see ${CLAUDE_SKILL_DIR}/scripts/suggestion.py --help). Inform the user that a suggestion was created for the global KB. 0a. Skill folder check: If topic has skill/ subfolder, route by content type: - "What/where/when" → KB concept notes (lightweight: summary, sources) - "How/when to use/procedures" → skill's reference/ or SKILL.md - Move overlapping "how" sections from concept notes into skill (no duplication) - No skill/ → all content to concept notes

  1. Templates from assets/: topic-note, index-note, legacy-note
  2. Slim frontmatter:

- Topic/index notes: only name + description in frontmatter. No title, summary, confidence, tags. - Inline <conf:high>, <conf:medium>, <conf:low> markers on claims are allowed and preserved during edits. Claims without a confidence tag should be treated as <conf:medium> (unverified but not flagged). - Body: # {Title} → content sections. No summary paragraph duplicating description. - TOC: Notes >100 lines need ## Contents in first 10 lines of body. - Legacy notes: name (append " (legacy)") + description (mention replacement).

  1. Keep heading accurate after modifications.
  2. Cross-references: markdown links with relative paths. Every concept note should link to at least one other note/index. Cross-KB: use @kb-name/path format. Global KBs: use soft references only — see: @namespace.kb/topic (no hard markdown links to read-only KBs).
  3. Index updates: Link new notes from relevant index. Create index if 3+ notes and none exists. Use table when section has 3+ items (see reference/folder-structure.md).

5a. Sibling unification: Skip if §4a already ran. Otherwise, if creating a new folder, check siblings per §4a.

  1. MANDATORY changelog: Every change to KB files → prepend entry to that KB's CHANGELOG.md. Follow reference/changelog-workflow.md. Do NOT skip.
  2. Large notes: If note reaches ~500 lines, suggest /kb-compact <path>.

4a. Shared Procedure: Structure Health Check

Run after any KB update. Reuse KB root listing from step 5 if available; otherwise one ls.

Detection: Group KB root folders by first hyphen-delimited segment. 3+ folders sharing prefix (no existing parent) = candidate. No candidate = stop.

If found: Ask user — "Restructure now", "Skip", "Not applicable".

If approved: Follow reference/folder-structure.md. Run validate_kb.py --quiet to confirm zero errors.

5. Shared Procedure: User Approval

Before non-trivial KB changes, present summary: notes to create, modify, index updates, changelog entry. Group claims by confidence level. Show <conf:low> claims in a separate "⚠ Low confidence" section. Use the AskUserQuestion tool (not plain text) with options: "Apply high/medium confidence", "Apply all (including low confidence)", "Apply selectively", "Cancel".

Exception: Minor changes (adding source URL, fixing typo) may proceed without approval but must be mentioned.

6. Note Naming and Folder Conventions

  • Lowercase, hyphen-separated. Topic-prefixed when ambiguous (crdt-operation-based.md, not operations.md). 2-4 words.
  • Index: one index.md per folder. See reference/folder-structure.md.

Skill Folders (KB-backed skills)

Topics with "how" content have skill/ inside their concept folder. Symlink from .claude/skills/ makes it discoverable. All skills must be KB-backed. When skill/ exists, concept notes are lightweight (provenance, summary, pointers to skill).

Forward-Only

Conventions apply to new notes by default. Existing notes not moved unless requested. When editing, may normalize frontmatter and update indexes.

Moving files

When relocation requested: mv files → run validate_kb.py → fix reported broken links. Don't rewrite content just to relocate.

7. Validation

After completing KB changes, run the validator to catch broken links, frontmatter issues, and structural errors:

uv run ${CLAUDE_SKILL_DIR}/scripts/validate_kb.py --quiet --json

Fix any reported errors before finishing.

平台分布

Codex

34.83%
按下载量换算24

Claude

33.87%
按下载量换算24

Cursor

17.58%
按下载量换算12

Gemini CLI

10.38%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/farzadshbfn/knowledge-skills --skill kb-learn 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills