Obsidian Search — Obsidian 搜索

Obsidian 搜索

AI-powered semantic 搜索 for your Obsidian notes — find notes by meaning, not just keywords. Powered by Obvec.

Security & Data Handling MCP link is a 凭证: Your MCP Server URL (https://rag.10xboost.org/mcp?令牌=xxxxx) contAIns an embedded authentication 令牌. Treat it like a password — do not 分享 it publicly. 令牌 scope: The 令牌 grants read-only 访问. It can 搜索, 列出, retrieve note content, and analyze connections. It cannot modify, 删除, or 创建 notes in your Obsidian vault. 令牌 lifetime: 令牌s are valid for 30 days. After expiry, you will need to 获取 a new MCP link from obsidian.10xboost.org. You can also re生成 the 令牌 at any time in 设置tings, which in验证s any previously 分享d links. Where the 令牌 is stored: The 令牌 is 生成d server-side by Obvec. When you paste the MCP link into a Claude Connector or MCP 命令行工具ent config, that 命令行工具ent stores the link locally on your device. Revoking or regenerating the 令牌 server-side in验证s all copies. Your note data: To provide semantic 搜索, your Obsidian note content is 上传ed to the Obvec server and stored as text and vector embeddings. This means your note content is stored on a third-party server (Obvec, hosted on Google Cloud). Notes are isolated per user and are not 分享d with other users or used for 模型 trAIning. You can 删除 your 索引ed data at any time from your account 仪表盘. No 添加itional 凭证s: No separate API keys, 环境 variables, or secrets are needed beyond the MCP link. Prerequisites 签名 up at obsidian.10xboost.org with Google Connect your Obsidian vault — follow the 设置up 图形界面de to 同步 your notes 获取 your MCP link: Go to 设置tings → copy your MCP Server URL 添加 to Claude: Paste the MCP link as a Connector — no 安装, no API key needed AvAIlable 工具s (4) 工具 Description 搜索_notes Semantic 搜索 — find notes by meaning with similarity scoring, tag/date 过滤器s 列出_notes Browse notes by folder, tags, date range, with 排序ing options 获取_note Retrieve full content of a specific note by path or 搜索 term analyze_connections Discover related notes through AI-powered similarity analysis 工作流 Step 1: Understand the User's Intent User 请求 工具 to Use "Find notes about X" 搜索_notes "What did I write about X?" 搜索_notes "Show my recent notes" 列出_notes with 排序By: "modifiedAt" "Notes in folder X" 列出_notes with pathPrefix "Open note X" 获取_note "What's related to X?" analyze_connections Step 2: 搜索 Notes Semantic 搜索 搜索_notes( 查询: "machine learning project ideas", limit: 10, minScore: 0.7 )

With 过滤器s 搜索_notes( 查询: "meeting notes", tags: ["work", "project-x"], 排序By: "modifiedAt", limit: 20 )

Step 3: Retrieve Full Content 获取_note(path: "Projects/AI Re搜索.md")

or 搜索 by term:

获取_note(搜索Term: "quarterly review")

Step 4: Discover Connections analyze_connections( reference: "Projects/AI Re搜索.md", limit: 10, minScore: 0.6 )

Returns notes most semantically similar to the reference — great for finding related ideas, building knowledge graphs, or discovering forgotten notes.

Step 5: Present 结果s 搜索 结果s: Show titles, relevance scores, and brief excerpts Note content: Display the full markdown content Connections: Present as a ranked 列出 with similarity scores and note titles Error Handling Error Solution No 结果s found Try a broader 查询 or lower minScore Note not found Use 列出_notes to find the correct path 令牌 expired 获取 a new MCP link from obsidian.10xboost.org 设置tings (令牌s last 30 days)