Hjarni
@hjarni
Hjarni について
Give your AI long-term memory. Hjarni is a note-taking app with a built-in MCP server. Claude and ChatGPT can search, read, create, and organize your notes. Markdown, folders, tags, and custom AI instructions per folder.
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"hjarni": {
"url": "https://hjarni.com/mcp",
"type": "streamable-http"
}
}
}ツール
24Get an overview of the Second Brain: counts of notes, containers, tags, inbox items, and recent notes.
Unified search across notes, containers, and tags. Returns results grouped by type. Use the types parameter to search only specific types (e.g., types: ['notes'] for note-only search). Supports full-text search with snippets, all note filters (container, tags, include_nested, search_scope), and archived note search via scope parameter.
List notes with optional filtering and sorting. Returns paginated results. Use team_id to list notes in a team. Use container_id to list notes in a specific container (with include_nested for sub-containers). Use tag/tags/tag_ids to filter by tags.
Get a single note by ID, including its full body content, tags, container, linked notes, and file attachments.
Create a new note. The body supports Markdown and wiki-links: use [[id:Note Title]] to link to other notes. Use team_id to create in a team.
Update an existing note. Supports updating content, moving to a container, changing tags, archiving (archived: true/false), and favoriting (favorited: true/false).
Permanently delete a note.
List containers (folders) for organizing notes. By default returns root-level personal containers. Use team_id to list containers in a team.
Get a single container by ID, including its LLM instructions if set. Use include_tree to also get ancestors and children.
Create a new container (folder) for organizing notes.
Update an existing container (rename, change description, move to a different parent, or set display position).
List all tags.
Create a new tag.
List all teams the user is a member of.
Get team details including recent notes.
Get LLM instructions at the specified level. Use level 'brain' to get global instructions (call early in conversations). Use 'personal_root' for personal space, 'container' for a specific container (includes inheritance chain), or 'team' for team instructions.
Update LLM instructions at the specified level. Use 'brain' for global, 'personal_root' for personal space, 'container' for a specific container, or 'team' for team instructions.
Create or remove a bidirectional link between two notes.
Attach a file to a note. Send the file contents as a base64-encoded string.
Fetch a file from a URL and attach it to a note. Follows one redirect.
Remove a file attachment from a note.
Generate a one-time upload URL for attaching a file to a note. Share this URL with the user so they can upload their file directly — the file bypasses the conversation and goes straight to Second Brain, saving tokens. The link expires after 30 minutes. Use files-check_upload to verify the upload completed.
Check the status of a file upload link created by files-create_upload_url. Returns 'pending' (user hasn't uploaded yet), 'completed' (file attached successfully), or 'expired' (link timed out).
Get a temporary download URL for a file attached to a note. Share the URL with the user so they can download the file in their browser. The URL expires after a few minutes.
概要
What is Hjarni?
Hjarni is a note-taking app with a built-in MCP server that gives your AI long-term memory. Notes are written in Markdown, organized in folders, and can include custom AI instructions per folder, allowing the AI to search, read, create, and organize notes without copying context into every conversation.
How to use Hjarni?
Connect your AI client to Hjarni using OAuth authentication—no API keys or Docker setup required. Use the MCP server to let your AI search, read, create, and organize notes. The free tier includes 25 notes with full MCP access.
Key features of Hjarni
- Markdown notes with wiki-links
- Folders with custom AI instructions
- Full-text search across all notes
- Bidirectional note linking
- Team collaboration with shared knowledge bases
- File attachments
Use cases of Hjarni
- Give your AI persistent access to a personal knowledge base.
- Provide per-folder behavior instructions to your AI for different contexts.
- Collaborate with a team and let the AI search and organize shared notes.
- Enable your AI to create and link new notes without manual copy-pasting.
FAQ from Hjarni
How does Hjarni compare to alternatives?
—
What are the runtime requirements?
Hjarni uses OAuth authentication and does not require API keys or Docker setup.
Where does my data live?
—
Are there any known limits?
The free tier includes 25 notes with full MCP access. Pro and Team tiers offer higher limits.
What transports or authentication does Hjarni use?
Hjarni uses OAuth authentication.
「メモリとナレッジ」の他のコンテンツ
Solomd
zhitongblogA markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.
Obsidian MCP Server
StevenStavrakisA simple MCP server for Obsidian
Obsidian MCP Server
cyanheadsRead, write, search, and surgically edit Obsidian vault notes, tags, and frontmatter via MCP. STDIO or Streamable HTTP.
Jupyter Notebook MCP Server (for Cursor)
jbenoModel Context Protocol (MCP) server designed to allow AI agents within Cursor to interact with Jupyter Notebook (.ipynb) files
Zettelkasten MCP Server
entanglrA Model Context Protocol (MCP) server that implements the Zettelkasten knowledge management methodology, allowing you to create, link, explore and synthesize atomic notes through Claude and other MCP-compatible clients.
コメント