Menü
知识库文档摄入工具。将文档加载、分块、嵌入并构建索引。 支持多种文件格式：Markdown、TXT、PDF、HTML。 支持增量更新和完整重建。
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
知识库分析工具。提供知识库统计、健康检查和优化建议。 分析文档分布、索引状态、查询性能等。
Location/geocoding tools using Open-Meteo Geocoding API
CRITICAL WEATHER TOOL. Use this skill IMMEDIATELY whenever the user asks about: - weather, temperature, forecast, rain, snow, sunny - e.g., "what is the weather in Hangzhou", "今天天气怎么样", "北京气温" - ALWAYS use this BEFORE attempting web search or other fallback tools. (Weather query tools using Open-Meteo with wttr.in fallback)
CRITICAL WEB SEARCH TOOL. Use this skill IMMEDIATELY whenever you need to search the web. IMPORTANT RULE: You MUST NOT use your native `search_web` tool. You MUST use this skill (`houyi_web_search`) instead via the Agent Execution Contract. Use this for: - Searching the internet for real-time information, news, or facts - Browsing websites and extracting content - Researching topics outside your training data (Multi-provider web search with caching, retry, and content extraction)
HouYi 项目设计文档与交付验收文档生成规范。 当需要为新特性创建设计文档（design doc）、验收跟踪文档（acceptance doc）， 或维护交付日志时，使用本 skill 提供的标准模板。 中文模板为基准版本（.CN.md），英文模板无后缀（默认）。
Task planning skill that creates and manages task plans in markdown files. Tracks progress through a structured plan file, ensuring tasks are completed systematically before marking as done.
| name | kb-ingest |
| version | 1.0.0 |
| description | 知识库文档摄入工具。将文档加载、分块、嵌入并构建索引。 支持多种文件格式：Markdown、TXT、PDF、HTML。 支持增量更新和完整重建。 |
| user-invocable | true |
| allowed-tools | ["Read","Write","Glob"] |
| hooks | {"PreToolUse":[{"matcher":"Write","type":"handler","handler":"houyi.rag.skills.kb_ingest.hooks:pre_ingest_hook"}],"PostToolUse":[{"matcher":".*","type":"handler","handler":"houyi.rag.skills.kb_ingest.hooks:post_ingest_hook"}],"Stop":[{"type":"handler","handler":"houyi.rag.skills.kb_ingest.hooks:stop_hook"}]} |
| invocationPolicy | {"modelAutoInvoke":"allow_with_consent","userInvocable":true,"sideEffect":"filesystem"} |
| permissions | {"filesystem":{"read":true,"write":true,"paths":["${KNOWLEDGE_DIR}/**/*","${WORKSPACE}/knowledge/**/*","${WORKSPACE}/.rag_index/**/*"]}} |
文档摄入工具，将文档加载、分块、嵌入并构建检索索引。
用户请求以下操作时激活：
| 参数 | 类型 | 默认值 | 说明 |
|---|---|---|---|
| paths | list[str] | 必填 | 要摄入的文件/目录路径 |
| knowledge_dir | str | "knowledge/" | 知识库根目录 |
| mode | str | "incremental" | 摄入模式：incremental/full |
| build_graph | bool | false | 是否构建知识图谱 |
| chunk_size | int | 512 | 文档分块大小 |
| chunk_overlap | int | 64 | 分块重叠大小 |
| 字段 | 类型 | 说明 |
|---|---|---|
| success | bool | 是否成功 |
| documents_processed | int | 处理的文档数 |
| chunks_created | int | 创建的分块数 |
| index_path | str | 索引存储路径 |
| message | str | 状态消息 |
扫描文件
加载文档
分块处理
嵌入生成
索引构建
持久化
用户: 将 docs/ 目录添加到知识库
Skill: 扫描 docs/ 目录，处理所有支持的文件
输出: 成功摄入 25 个文档，创建 342 个分块
{
"type": "object",
"properties": {
"paths": {
"type": "array",
"items": { "type": "string" },
"description": "Files or directories to ingest"
},
"knowledge_dir": {
"type": "string",
"default": "knowledge/",
"description": "Knowledge base directory"
},
"mode": {
"type": "string",
"enum": ["incremental", "full"],
"default": "incremental"
},
"build_graph": {
"type": "boolean",
"default": false
},
"chunk_size": {
"type": "integer",
"default": 512
},
"chunk_overlap": {
"type": "integer",
"default": 64
}
},
"required": ["paths"]
}
{
"type": "object",
"properties": {
"success": { "type": "boolean" },
"documents_processed": { "type": "integer" },
"chunks_created": { "type": "integer" },
"index_path": { "type": "string" },
"message": { "type": "string" }
}
}