data-cleaning
Skill0 starsChanged 3 months agoMITView on GitHub
为 LLM / RAG / 微调准备干净、高质量的数据。当需要清洗、去重、去噪、去隐私、标准化文本数据,或检查数据集质量问题时使用。
8 other projects use a version of this.
Adopted by
Most recently taken up by
- SSameer20/data-analyst-toolkittheir own version, first seen Nov 8, 2026
- genspark-ai/genofficetheir own version, first seen Oct 25, 2026
- cellrishi-code/claude-skills-atlastheir own version, first seen Oct 4, 2026
- magnus919/agent-skillstheir own version, first seen Jul 21, 2026
- qunyou-agent/finance-skillstheir own version, first seen Jul 6, 2026
- minagayid/Agentstheir own version, first seen Jul 5, 2026
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 8
- genspark-ai/genoffice★ 12ksame name
- magnus919/agent-skills★ 172same name
- claude-office-skills/claude-office-plugin★ 9same name
- cellrishi-code/claude-skills-atlas★ 6same name
- qunyou-agent/finance-skills★ 4same name
- SSameer20/data-analyst-toolkit★ 4same name
- minagayid/Agents★ 1same name
- adelcreative030-debug/Agent-Skills★ 0same 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 repository5 other files it gives its agents.
- fine-tuning-prepSkill
- llm-evaluationSkill
- prompt-engineeringSkill
- prompt-optimizerSkill
- rag-builderSkill
The file
Files
面向 LLM 的数据清洗
数据质量决定 RAG 检索与微调效果的上限。本技能提供一套可复用的清洗与质检流程。
使用步骤
1. 抽样了解数据(先看样貌)
- 随机抽样 + 分布统计:长度、语言、编码、字段缺失率。
- 记录典型噪声:乱码、格式混乱、重复、HTML/标签残留、敏感信息。
2. 分维度清洗
- 文本规范化:统一换行/空白、Unicode 归一化(NFC)、去除不可打印字符、统一引号。
- 去噪:去掉广告/导航/模板页眉脚(对爬取内容尤其重要)、HTML/XML 标签、Markdown 符号按需保留或剥离。
- 去重:按指纹或归一化文本做近似去重(如 minhash / simhash / embedding 相似度),保留代表样本。
- 去敏感(隐私):识别并打码/删除身份证、手机号、邮箱、密钥等(用正则 + 关键词 + 可选的 NER)。
- 语言/粒度:按需过滤非目标语言、太短片段、异常长度(过长截断、过短丢弃)。
3. 抽取有效内容(面向 RAG/微调)
- 按标题/章节切片为语义完整的块。
- 保留元数据(来源、时间、标题),便于追溯与检索。
- 对表格/代码/公式单独标记类型,避免混排泄漏。
4. 质检与量化(清洗前 → 清洗后)
对每个维度统计清洗前后对比,形成质检报告:
样本总数 / 去重去掉 / 去噪去掉 / 脱敏条数 / 保留率抽样人工复核若干条,确保清洗没有误删有效信息。
5. 输出
- 输出干净数据集 + 元数据清单 + 质检报告。
- 建议同步输出清洗脚本(可复现),并保留原始数据以便追溯。
自查清单
- 是否先抽样了解再清洗(避免一刀切误删)。
- 去重/去噪的阈值是否合理,是否覆盖代表性样本。
- 隐私信息是否已脱敏。
- 是否保留可复现的清洗脚本。
- 清洗前后是否有量化对比与人工抽样复核。
参考关联
- 清洗后的数据用于 RAG 见 rag-builder;用于微调见 fine-tuning-prep。
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
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.

