Lokaler MCP in Node.js + TypeScript, um Ihren Obsidian-Vault für VS Code + GitHub Copilot verfügbar zu machen.
Er wurde für die lokale Nutzung via stdio entwickelt, mit Fokus auf das Lesen und Schreiben von Markdown-Notizen innerhalb Ihres Vaults.
npm install -g obsidian-mcp-local
Related MCP server: Obsidian MCP Server
search_notes(query)
sucht nach Text im Pfad, Frontmatter und Inhalt der Notizen
gibt gerankte Ergebnisse mit einem kurzen Ausschnitt zurück
get_note(path)
öffnet eine Notiz aus dem Vault
gibt path, frontmatter und content zurück
create_note(path, content, overwrite?)
erstellt eine neue Notiz
überschreibt optional eine bestehende Notiz
append_to_note(path, content)
fügt Inhalt am Ende einer bestehenden Notiz hinzu
find_by_tag(tag)
findet Notizen nach Tag
unterstützt tags im Frontmatter und Inline-Tags im Inhalt
greift nur auf Dateien innerhalb des konfigurierten Vaults zu
ignoriert Verzeichnisse wie:
.obsidian
.git
node_modules
arbeitet nur mit .md-Dateien
normalisiert Pfade, um Zugriffe außerhalb des Basisverzeichnisses zu verhindern
obsidian-mcp-local/
package.json
tsconfig.json
README.md
.vscode/
mcp.example.json
src/
index.tsNode.js 20+
npm
VS Code mit GitHub Copilot
ein lokaler Obsidian-Vault
Im Projektverzeichnis:
npm install
npm run buildFür die Entwicklung:
npm run devUm die kompilierte Version auszuführen:
npm startnpm install
npm run buildKopieren Sie den Inhalt von .vscode/mcp.example.json in Ihre .vscode/mcp.json im Workspace, in dem Sie Copilot verwenden möchten.
Beispiel: bei lokaler Nutzung:
{
"servers": {
"obsidian-local-vault": {
"type": "stdio",
"command": "node",
"args": ["C:/caminho/para/obsidian-mcp-local/dist/index.js"],
"env": {
"OBSIDIAN_VAULT_PATH": "D:/Obsidian/Vault"
}
}
}
}bei Nutzung via npx (kein lokaler Build erforderlich):
{
"servers": {
"obsidian-local-vault": {
"command": "npx",
"args": ["-y", "obsidian-mcp-local"],
"env": {
"OBSIDIAN_VAULT_PATH": "D:/Obsidian/Vault"
}
}
}
}Ersetzen Sie:
C:/caminho/para/obsidian-mcp-local/dist/index.js
D:/Obsidian/Vault
durch die tatsächlichen Pfade auf Ihrem Rechner.
Danach sollte Copilot den MCP-Server erkennen.
„Suche in meinem Vault nach Notizen über .NET“
„Öffne die Notiz knowledge/backend/dotnet.md“
„Erstelle eine Notiz in inbox/ideias-mcp.md mit einer Zusammenfassung dessen, was wir besprochen haben“
„Füge am Ende der Notiz daily/2026-04-06.md den Text - testar MCP local hinzu“
„Finde Notizen mit dem Tag #arquitetura“
append_under_heading
Parsing von [[wikilinks]]
get_backlinks(note)
SQLite-Index für schnelle Suche
Whitelist von Ordnern für Schreibzugriffe (inbox/, daily/, scratch/)
konfigurierbare Schreibsperre für bestimmte Ordner
Dieses Projekt hängt nicht davon ab, dass Obsidian geöffnet ist.
Es arbeitet direkt auf den Dateien des Vaults.
Wenn Sie Schreibzugriffe sowohl in Obsidian als auch in VS Code aktivieren, liegt die Kontrolle über die Nebenläufigkeit bei Ihnen.
Das Projekt geht derzeit davon aus, dass der Vault ein lokaler Markdown-Ordner ist.
Die Implementierung befindet sich in:
src/index.ts
Persönliche Nutzung / Basis für Anpassungen.
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | ||
| content | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | ||
| content | Yes | ||
| overwrite | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| tag | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.2append_to_notecreate_notefind_by_tagget_notesearch_notesScored across 5 tools
Each tool name clearly indicates a distinct operation: create, get, append, search, and find by tag. No significant overlap is apparent.
All tool names follow a consistent verb_noun pattern in snake_case (e.g., create_note, search_notes), making the set predictable.
Five tools cover essential note operations (create, retrieve, modify, search, tag) without unnecessary bloat or deficiency for the domain.
The set includes create, read, update (via append), and search operations but lacks a delete operation, which is a notable gap for a complete CRUD surface.
Search and reason over your Obsidian-style Markdown vault, right from ChatGPT.
Connect AI assistants to your GitHub-hosted Obsidian vault to seamlessly access, search, and analy…
Search your Obsidian vault to quickly find notes by title or keyword, summarize related content, a…
Personal context for every AI: search, read, and write back to your private Markdown library.
