{"id":"150e329d-05fd-4307-8b2e-2f34ae5a3adf","slug":"clawhub-ldxs001-local-rag-builder","name":"local-rag-builder","description":"本地 RAG 系统搭建技能，支持环境检测修复、嵌入模型多源下载、5种切分策略 + GuardStack + 后处理 + 插件注册、多知识库管理 + 自动分类规则、可调 Prompt、Web 可视化配置 + 极客模式 + 模板管理","canonicalUrl":"https://www.xpersona.co/agent/clawhub-ldxs001-local-rag-builder","sourceUrl":"https://clawhub.ai/ldxs001/local-rag-builder","homepage":"https://clawhub.ai/ldxs001/skills/local-rag-builder","source":"CLAWHUB","vendor":{"slug":"clawhub","label":"Clawhub","url":"https://clawhub.ai/ldxs001/skills/local-rag-builder"},"protocols":["OPENCLEW"],"capabilities":[],"trustScore":null,"trustConfidence":"unknown","artifactCount":0,"benchmarkCount":0,"lastRelease":"1.6.0","freshnessAt":"2026-10-11T03:18:52.066Z","freshnessLabel":"Oct 11, 2026","securityReviewed":true,"openapiReady":false,"stats":[{"label":"Trust score","value":"Unknown"},{"label":"Compatibility","value":"OpenClaw"},{"label":"Freshness","value":"Oct 11, 2026"},{"label":"Vendor","value":"Clawhub"},{"label":"Artifacts","value":"0"},{"label":"Benchmarks","value":"0"},{"label":"Last release","value":"1.6.0"}],"factsPreview":[{"factKey":"vendor","category":"vendor","label":"Vendor","value":"Clawhub","href":"https://clawhub.ai/ldxs001/skills/local-rag-builder","sourceUrl":"https://clawhub.ai/ldxs001/skills/local-rag-builder","sourceType":"profile","confidence":"medium","observedAt":"2026-10-11T03:18:52.138Z","isPublic":true},{"factKey":"protocols","category":"compatibility","label":"Protocol compatibility","value":"OpenClaw","href":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-local-rag-builder/contract","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-local-rag-builder/contract","sourceType":"contract","confidence":"medium","observedAt":"2026-10-11T03:18:52.138Z","isPublic":true},{"factKey":"traction","category":"adoption","label":"Adoption signal","value":"1.2K downloads","href":"https://clawhub.ai/ldxs001/local-rag-builder","sourceUrl":"https://clawhub.ai/ldxs001/local-rag-builder","sourceType":"profile","confidence":"medium","observedAt":"2026-10-11T03:18:52.138Z","isPublic":true},{"factKey":"latest_release","category":"release","label":"Latest release","value":"1.6.0","href":"https://clawhub.ai/ldxs001/local-rag-builder","sourceUrl":"https://clawhub.ai/ldxs001/local-rag-builder","sourceType":"release","confidence":"medium","observedAt":"2026-07-11T04:49:29.206Z","isPublic":true},{"factKey":"handshake_status","category":"security","label":"Handshake status","value":"UNKNOWN","href":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-local-rag-builder/trust","sourceUrl":"https://www.xpersona.co/api/v1/agents/clawhub-ldxs001-local-rag-builder/trust","sourceType":"trust","confidence":"medium","observedAt":null,"isPublic":true}],"highlights":["1.2K downloads","Trust evidence available"],"agentCard":{"name":"local-rag-builder","description":"本地 RAG 系统搭建技能，支持环境检测修复、嵌入模型多源下载、5种切分策略 + GuardStack + 后处理 + 插件注册、多知识库管理 + 自动分类规则、可调 Prompt、Web 可视化配置 + 极客模式 + 模板管理","source":"CLAWHUB","sourceId":"clawhub:s1794v2r46s8y5d1r4jdd7ec5h84rw70:local-rag-builder","homepage":"https://clawhub.ai/ldxs001/skills/local-rag-builder","repository":"https://clawhub.ai/ldxs001/local-rag-builder","documentation":"https://www.xpersona.co/agent/clawhub-ldxs001-local-rag-builder","protocols":["OPENCLEW"],"examples":[{"kind":"example","language":"python","snippet":"model_path = \"D:/models/bge-small-zh-v1.5\""},{"kind":"example","language":"text","snippet":"File v1.6.0:references/faq.md\n\n# 常见问题 (FAQ) — local-rag-builder\r\n\r\nQ: 为什么 Python 3.12 无法安装 chromadb？\r\nA: chromadb 的 Windows 预编译轮子最高支持到 Python 3.11。建议使用 conda 创建 3.11 虚拟环境。\r\n\r\nQ: 模型下载失败怎么办？\r\nA: 本工具内置 4 个下载源（ModelScope、HuggingFace 镜像、官方源、LLM 搜索），每个源会自动重试 3 次。如果全部失败，可以尝试：\r\n1. 配置环境变量 `HF_ENDPOINT` 为 `https://hf-mirror.com` 后重试\r\n2. 使用 `--interactive` 模式选择其他源\r\n3. 手动下载后使用 `--check` 验证\r\n\r\nQ: 多个知识库如何切换？\r\nA: 在 CLI 中使用 `/kb use <name>` 命令，或在配置文件的 `kb.active_kb` 字段指定。\r\n\r\nQ: Prompt 模板如何持久化？\r\nA: Prompt 模板保存在 `data/prompts/custom_prompt_template.txt`，程序重启后自动加载。使用 `/prompt set` 或 `--set` 命令配置后即持久化。\r\n\r\nQ: 如何重置所有配置恢复到初始状态？\r\nA: 运行 `python -c \"from config import reset_config; reset_config()\"` 或从 Web 界面点击\"重置配置\"按钮。这会清除 `data/config/rag_config.json` 并恢复默认值，同时重置 Prompt 模板。注意：重置不会删除知识库数据和已下载的嵌入模型。\r\n\r\nQ: 本技能和本地 LLM（如 LM Studio）是什么关系？\r\nA: 本技能本身可以扮演 LLM 角色，但如果你有本地运行的 LM Studio / Ollama 等服务，也可以通过配置 LLM 地址接入。技能自动适配两种模式。\r\n\r\nQ: 向量的相似度阈值如何配置？\r\nA: 在检索配置中配置 `score_threshold`（0-1 之间的浮点数），设为 `null` 则不启用阈值过滤。\r\n\r\nQ: Windows 上模型路径名变形如何处理？\r\nA: ModelScope 下载的模型名中 `.` 可能变为 `___`（如 `bge-small-zh-v1___5`）。本工具会自动检测并修正路径，无需手动处理。\n\nFile v1.6.0:references/guide.md\n\n# 使用指南 — local-rag-builder\r\n\r\n本指南提供 local-rag-builder 的完整使用教程，从环境搭建到高级配置。\r\n\r\n---\r\n\r\n## 目录\r\n\r\n1. [快速入门](#快速入门)\r\n2. [环境检测与安装](#环境检测与安装)\r\n3. [嵌入模型下载与管理](#嵌入模型下载与管理)\r\n4. [文本切分配置](#文本切分配置)\r\n5. [知识库管理](#知识库管理)\r\n6. [Prompt 自定义](#prompt-自定义)\r\n7. [Web 界面配置](#web-界面配置)\r\n8. [两种运行模式](#两种运行模式)\r\n9. [技能模式：智能体接口](#技能模式智能体接口)\r\n10. [独立模式：外部 LLM 接入](#独立模式外部-llm-接入)\r\n11. [故障排除](#故障排除)\r\n\r\n---\r\n\r\n## 快速入门"}]}}