agentleFS
JingHao-LeonJingHao-Leon/fde-case-lab/skills/kb-build/SKILL.md

kb-build

Skill0 starsChanged 2 months agoMITView on GitHub

从企业文档与专家隐性经验构建知识库,并把真实高频问题变成测试集做验收(Recall 验收线 ≥80%)。当用户要"沉淀专家经验""建知识库/资料库"时使用。

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. luanmorenommaciel/agentspec★ 256same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

Files
SKILL.md

KB Build · 经验知识库构建与验收

对应案例:NO.1(制造经验传承)、NO.11(外贸资料资产化)。 参考实现:cases/01-knowledge-base/(词表归一化 + BM25/TF-IDF 混合检索, Recall@5 99.2%,带噪 97.1%)。

何时使用

  • 企业要沉淀专家隐性经验、散落文档,让"新人能查到"
  • 已有知识库但"查不到、不敢用",需要重建检索质量
  • 需要为知识库定义可量化验收标准

流程

  1. 高频问题收集 = 测试集:访谈/工单/群聊里提取真实高频问题,逐条标注 gold 答案片段——测试集在建库前就定,验收标准与客户共定(默认 Recall@5 ≥ 80%)。
  2. 语料入库与解析率检查:OCR/解析后抽查解析质量(案例 1 口径约 95%), 低于阈值先修解析再谈检索。
  3. 词表归一化(第一杠杆):收集同一实体的不同叫法(别名/口语/术语), 建别名→规范词映射,query 与文档同侧归一化。实测证明这比换检索器重要 (66.2%→99.2% vs 换检索器反而 55.8%)。
  4. 混合检索:BM25(char-bigram) + TF-IDF(char 2-4 gram) 融合; 同前缀冗余高时加 MMR 重排(λ=0.7)。
  5. 人机验收:跑测试集出指标;不过线先修词表,再修分块,最后才考虑换模型。
  6. 接入现有流程:知识库入口嵌进用户已有工具(培训平台/IM),不做孤立入口。

验收

  • 测试集 Recall@5 ≥ 80%(与客户共定的线)
  • 带噪语料(模拟解析损失)下 Recall@5 不塌方
  • 每条检索结果可追溯到源文档

边界

  • LLM 生成答案环节不在本技能内(那是 RAG 应用层);本技能保证"检索得准"
  • 涉密的隐性经验是否入库,由专家本人确认

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.