Skip to content
cccskills
FreePublished on GitHub

knowledge-base

Search configured Yuxi knowledge bases to answer questions using internal or uploaded documents. Open source text, locate relevant passages, inspect mind maps, and identify supporting files so answers can be traced back to their evidence.

OriginalShow the original description

使用 Yuxi 知识库进行检索、打开文档、文档内定位和查看思维导图。当用户需要基于已配置知识库回答问题、核验资料或引用文档内容时使用此技能。

See how to install

Good for

  • Answering questions from internal documents
  • Locating passages that support a claim
  • Exploring knowledge base structure with mind maps
  • Retrieving original files for further analysis

Files included (1)

  • SKILL.md2.7 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

知识库技能

当用户要求基于项目知识库、内部资料、上传入库文档或知识图谱相关内容回答问题时,使用此技能。

可用工具

  • list_kbs:列出当前会话可访问且已启用的知识库。
  • query_kb:按 kb_id 在指定知识库中检索内容,返回 file_id 和相关片段。
  • open_kb_document:按 kb_id 和 file_id 打开文档原文窗口,适合查看更完整上下文。
  • find_kb_document:在已知文档内用关键词或正则定位段落。
  • get_mindmap:查看知识库思维导图结构。
  • search_file:按文件名关键词搜索知识库中的文件,支持指定知识库或跨知识库,返回文件列表与分页信息。
  • download_kb_file:按 kb_id 和 file_id 下载知识库原始二进制(pdf/docx/xlsx 等)到沙盒 outputs 目录,返回沙盒内可见的 virtual_path。当后续需要用代码读取原始文件结构(如 openpyxl 读 xlsx 单元格、pdfplumber 重新解析版面)时使用;query_kb/open_kb_document 只返回文本切片,无法满足需要文件对象的场景。

操作流程

  1. 需要先确认当前会话有哪些知识库可用;不确定时调用 list_kbs。
  2. 针对用户问题选择最相关的知识库,使用 query_kb 检索。
  3. 检索结果的 metadata.chunk_count == 1 表示文件只有一个分片;除非需要原文行号或额外上下文,否则无需打开。大于 1 且当前片段不足以回答时,使用返回的 file_id 调用 open_kb_document 查看上下文。字段缺失或为 0 表示未知,不据此判断文档完整。
  4. 如果用户要求定位术语、指标、章节或原文证据,使用 find_kb_document 在候选文档内查找。
  5. 当用户关心知识库结构、文件分类或知识框架时,使用 get_mindmap。

关键约束

  • 只能访问当前会话配置和用户权限允许的知识库。
  • 不要编造 kb_id 或 file_id;优先从 list_kbs 和 query_kb 的返回结果中获取。
  • 回答需要可追溯时,应说明依据来自哪个知识库、文件或检索片段。
  • Dify 等外部只读知识库可能只支持检索,不一定支持打开全文或文档内查找;遇到工具返回限制说明时,应如实告知用户。

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

deep-research

FreeSummary

Organize complex research by splitting a topic into focused questions, assigning research agents, and checking important claims. Findings become a structured report with source links, conflicting evidence, and clearly stated uncertainties.

  • Reviewing research literature
  • Researching industries and competitors
  • Comparing technologies for selection
xerrors/Yuxi7.4KUpdated Oct 17, 2026

html-preview

FreeSummary

Add compact visual summaries to Markdown answers using static HTML and CSS. Show comparisons, key metrics, timelines, or simple relationships in a preview, while keeping explanations and detailed information in the surrounding text.

  • Comparing key metrics visually
  • Showing a process or sequence of steps
  • Summarizing events on a timeline
xerrors/Yuxi7.4KUpdated Oct 17, 2026

image-gen

FreeSummary

Generate images, posters, and illustrations from written descriptions using an external image API, with Qwen-Image as the default. Save the finished image in the agent's working environment and display it through the artifact presentation tool.

  • Generating an illustration from a written description
  • Creating a poster with a specified style
  • Saving and displaying images generated with Qwen-Image
xerrors/Yuxi7.4KUpdated Oct 17, 2026

mysql reporter

FreeSummary

Query MySQL databases to summarize sales, user behavior, and business metrics. The skill checks available tables, runs read-only queries, and presents results as Markdown tables with explanations of key figures, changes, and conclusions.

  • Summarizing sales data
  • Analyzing user behavior
  • Reporting business metrics
xerrors/Yuxi7.4KUpdated Oct 17, 2026

See all skills by xerrors

Report a problem with this skill