Sunub Obsidian Mcp Server
Connects to your local Obsidian vault and exposes hybrid search that combines keyword matching with local semantic embeddings using transformers.js. The vault tool provides actions for searching, reading notes, collecting context around topics, and managing frontmatter properties. Search uses RRF fusion and BGE reranking models that run locally without external API calls. Also includes a terminal based chat UI with RAG integration and slash commands to query your vault directly. You'll need to run the setup command to download the embedding and reranking models first. Reach for this when you want Claude to search and reason over your personal knowledge base without uploading anything to the cloud.
- GitHub stars
- 2
- tools
- 5
- auth
- Secret needed
- transport
- HTTP
- last updated
- Sep 18, 2025
Tools
5 toolsPublic tool metadata for what this MCP can expose to an agent.
vaultMCP tool with actions search, read, list_all, stats, collect_context, and load_memory.
MCP tool with actions search, read, list_all, stats, collect_context, and load_memory.
No parameter schema in public metadata yet.
generate_propertyGenerates frontmatter candidates from document content.
Generates frontmatter candidates from document content.
No parameter schema in public metadata yet.
write_propertyWrites frontmatter properties.
Writes frontmatter properties.
No parameter schema in public metadata yet.
create_document_with_propertiesTwo-step workflow that analyzes a document and then generates and writes its properties.
Two-step workflow that analyzes a document and then generates and writes its properties.
No parameter schema in public metadata yet.
organize_attachmentsOrganizes attachments within a document and updates links.
Organizes attachments within a document and updates links.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Obsidian MCP Server
obsidian-mcp-server는 Obsidian Vault의 Markdown 문서를 AI 에이전트가 조회하고, 관련 근거를 선별하고, 컨텍스트로 압축해 활용할 수 있게 해주는 로컬 우선 MCP 서버입니다.
이 프로젝트에서 RAG는 특정 벡터 DB나 검색 엔진 선택을 뜻하지 않습니다. RAG는 Vault에서 후보 문서를 찾고, Agent 작업에 맞는 근거를 고르고, 컨텍스트 윈도우에 맞게 압축해 제공하는 Agent Context Pipeline입니다.
핵심 방향
- 로컬 우선 동작: 외부 서비스나 별도로 운영해야 하는 검색 엔진 없이 사용자 머신에서 Vault 검색과 컨텍스트 준비를 수행합니다.
- 컨텍스트 선별: 키워드 검색과 시맨틱 검색을 결합해 Agent에게 제공할 근거 후보를 찾습니다.
- 명시적 RAG 통합: 모든 프롬프트에 백그라운드 검색을 붙이지 않고, Vault 관련 명령이나 도구 참조가 있을 때 문맥 수집 경로를 엽니다.
- 토큰 절약: 원문 전체를 밀어넣기보다 excerpt, evidence snippet,
memory_packet형태로 압축합니다. - 로컬 모델 기반 검색:
@huggingface/transformers, LanceDB, local reranker를 사용해 semantic retrieval을 로컬에서 수행합니다.
Elasticsearch 같은 검색 엔진도 retrieval backend로 사용할 수 있는 대안입니다. 다만 이 프로젝트의 기본 목표는 외부 서비스나 별도 검색 엔진에 의존하지 않는 로컬 단독 작업이므로, embedded retrieval stack을 기본값으로 선택합니다.
제공 기능
MCP Tools
Details
- Repository
- github.com/sunub/obsidian-mcp-server
- Categories
- Documents & Knowledge, Search & Web Crawling
- Transport
- HTTP
- Auth
- Required
- Registry
- MCP registry, active
Related Documents & Knowledge MCP servers
View all
Google Driveus-all/google-drive-mcp-serverGoogle Drive + Workspace MCP — Docs, Sheets, Slides, Shared Drives, Labels, Approvals, DWD1stars
NanoPDFr09722akabennett/nano-pdf-mcpAI-powered PDF editing and generation MCP server using Gemini Pro.1stars
Document Processingfilegraph/docconvertExtract text from documents, manipulate PDFs, and perform OCR on images.0stars
OntoRamp Knowledge Cartographermatthewdavisaiarchitect/ontoramp-knowledge-cartographerKnowledge coverage map and health score. Ingest docs into a governed knowledge graph via MCP.0stars