Obsidian MCP
Enables AI agents to manage Obsidian vaults through full CRUD operations, wikilink management, and section-level manipulation. It supports frontmatter editing, tag-based searching, and automated link updates to maintain vault integrity.
README
Obsidian MCP
MCP server for Obsidian vault management. Designed for AI agents (Claude Desktop, Claude Code).
Features
- Full CRUD operations on notes
- Wikilink management (auto-update on rename/delete)
- Frontmatter manipulation
- Tag-based and content search
- Backlinks discovery
- Section-level operations
Installation
poetry install
Configuration
Set the vault path via environment variable:
export OBSIDIAN_VAULT_PATH="/path/to/your/obsidian/vault"
# WSL example (Windows vault accessed from WSL)
export OBSIDIAN_VAULT_PATH="/mnt/c/Users/username/Documents/Obsidian/my_vault"
Claude Desktop Configuration
Add to claude_desktop_config.json:
Linux/macOS:
{
"mcpServers": {
"obsidian": {
"command": "bash",
"args": ["-c", "cd /path/to/obsidian-mcp && poetry run obs-mcp"],
"env": {
"OBSIDIAN_VAULT_PATH": "/path/to/your/obsidian/vault"
}
}
}
}
Windows (via WSL):
{
"mcpServers": {
"obsidian": {
"command": "wsl",
"args": [
"-e", "bash", "-c",
"cd /path/to/obsidian-mcp && poetry run obs-mcp"
],
"env": {
"OBSIDIAN_VAULT_PATH": "/mnt/c/Users/username/Documents/Obsidian/my_vault"
}
}
}
}
Available Tools
| Tool | Description |
|---|---|
read_note(name) |
Read note content (frontmatter + body) |
create_note(name, content?, frontmatter?) |
Create note |
append_note(name, text) |
Append to note |
update_note(name, content) |
Replace body, preserve frontmatter |
delete_note(name, dry_run?) |
Delete (links updated to [[name (deleted)]]) |
rename_note(old_name, new_name, dry_run?) |
Rename + update wikilinks |
search_notes(query, mode?) |
Search (name, name_partial, content, tag) |
get_links(name, direction?) |
Get incoming/outgoing links |
get_note_metadata(name) |
Get frontmatter + links without body |
set_frontmatter(name, key, value) |
Set frontmatter key |
add_tag(name, tag) |
Add tag to note |
remove_tag(name, tag) |
Remove tag from note |
list_notes(limit?, offset?) |
List all notes (paginated) |
replace_text(name, old, new, replace_all?) |
Replace text in note |
insert_text(name, text, before?, after?) |
Insert text at position |
read_section(name, section) |
Read section by heading |
append_section(name, section, text) |
Append to section |
update_section(name, section, content) |
Replace section content |
delete_section(name, section) |
Delete section |
get_headings(name) |
Get all headings |
find_broken_links() |
Find broken wikilinks |
get_help() |
Get tool descriptions |
Vault Structure
- Flat structure (all
.mdfiles in vault root) - YAML frontmatter with
---delimiters - Wikilinks:
[[Note Name]] - Tags in frontmatter:
tags: [tag1, tag2] .trash/for deleted notes.obsidian/ignored
Development
# Run tests
poetry run pytest tests/spec/core/ -v
# Lint
poetry run ruff check src/
poetry run mypy src/
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。