agentmods.dev

obsidian-mcp: Instructions file for Claude Code

CLAUDE.md

obsidian-mcp CLAUDE.md is an instructions file for Claude Code from maxkuminov/obsidian-mcp. It costs 13,411 tokens per session, scanned A, original, MIT.

Project instructions for an Obsidian MCP server that exposes a large collection of Obsidian notes through semantic search, full-text search, and guided exploration.

In plain words
What is it for?
Use them when developing, configuring, or maintaining this self-hosted Obsidian note-search server.
Why use it?
They explain the server's technology, layout, database, search, and embedding setup so an agent can work on the project with the right context.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions subagents.

This is maxkuminov/obsidian-mcp's own configuration. It tells Claude Code how to work on obsidian-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything obsidian-mcp configures →

not rated 18repo changed yesterday A scan 13,411 tokens original MIT
View source ↗ maxkuminov/obsidian-mcp
Reuse

Borrowing it

Nothing to install: this file belongs to maxkuminov/obsidian-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/maxkuminov/obsidian-mcp/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/maxkuminov/obsidian-mcp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for obsidian-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/maxkuminov/obsidian-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/maxkuminov/obsidian-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/maxkuminov/obsidian-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/maxkuminov/obsidian-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for obsidian-mcp CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/maxkuminov/obsidian-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/maxkuminov/obsidian-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 13,411 This file is loaded in full into every session.
When invoked 13,411 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.13411 $0.13411
Opus 5.5 $0.05364 $0.05364
Sonnet 5.5 $0.02682 $0.02682
Haiku 4.5 $0.01341 $0.01341

Measured yesterday against content hash ca0b82e0dd51, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

obsidian-mcp CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 724 lines

How it starts

The opening of the file, as written. The whole thing — 724 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Obsidian MCP Server

Self-hosted MCP server exposing an Obsidian vault (~2,577 markdown files) via semantic search, full-text search, and agentic exploration.

Stack

  • Python 3.12 / FastAPI / uvicorn
  • SQLAlchemy async + asyncpg + pgvector
  • MCP Python SDK (Streamable HTTP)
  • PostgreSQL 16 with pgvector (shared instance — see .env)
  • Embedding provider abstraction (src/services/embeddings.py):
    • OllamaProvider — bge-m3 by default, set OLLAMA_URL in .env
    • OpenAIProvider — text-embedding-3-{small,large} over httpx, supports Azure OpenAI / OpenAI-compatible base URLs
  • Jinja2 control panel; Chart.js vendored under src/control_panel/static/vendor/, no CDN, hand-written CSS, nonce-based CSP (PANEL_CSP) — see control panel.

Project Layout

  • src/main.py — FastAPI app, lifespan, MCP mount
  • src/config.py — pydantic-settings
  • src/database.py — async SQLAlchemy engine/session
  • src/models/db.py — ORM models (api_keys, usage_logs, notes_metadata, note_embeddings, note_links, indexer_runs, transfer_tokens, oauth_clients/codes/tokens)
  • src/mcp_server/ — MCP server, tools, auth middleware
  • src/services/ — vault ops, search, embeddings, indexer, transfer, anchored FS
  • src/transfer/ — public /transfer/* capability-redemption routes
  • src/api/ — control panel REST endpoints
  • src/control_panel/ — Jinja2 templates + static assets
  • alembic/ — database migrations

Infrastructure

  • Container: obsidian-mcp, listens on :8000
  • Traefik routes: hostname driven by MCP_HOSTNAME in .env
    • Panel routes: OAuth protected via chain-oauth@file
    • MCP routes (/mcp/*): API key auth at app level
    • Transfer routes (/transfer/*): same router as /mcp (no OAuth chain); capability-token auth at app level
  • Registry: localhost:5000 (or change in Makefile)
  • Production deploy: merge to main → CI → image workflow (GHCR, tag main-<sha>-<run_number>) → Flux image automation on the homelab k3s (maxkuminov/infra docs/homelab-image-automation.md). Until the app is cut over there, a digest-bump PR in maxkuminov/infra. Docs-only merges build nothing.
  • Release tag (v*, cut at a PR's merge commit) → registry-publish workflow: retags the sha- image CI proved as ghcr.io/maxkuminov/obsidian-mcp:<tag> (no rebuild) and publishes server.json to the official MCP Registry via GitHub OIDC. Not a deploy; the GHCR package must be public for the registry's OCI validation (owner setting).
  • make deploy (build → scan → push → backup → migrate → recreate) is the compose path, not production.

Read the full file on GitHub · 724 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday Changed · +30 lines · +645 tokens per session ca0b82e0dd51
  2. 3d ago Changed · +6 lines · +89 tokens per session d81e4ee00e2f
  3. 5d ago Changed · +82 lines · +1,714 tokens per session 4d32357e52d4
  4. 6d ago Changed · +103 lines · +1,968 tokens per session 10175a4ca471
  5. 7d ago Changed · +72 lines · +1,397 tokens per session ea08ae4d4eb9
  6. 13d ago Changed 6c037c341717
  7. 26d ago Changed · +39 lines · +734 tokens per session 443a961e0257
  8. 27d ago Changed · +27 lines · +566 tokens per session 21269c4582b0
  9. 1mo ago Changed · +17 lines · +310 tokens per session 4debd24b4190
  10. 1mo ago First seen · 348 lines · 5,988 tokens per session scan A e4e83482b465

Subscribe to this mod's changes

obsidian-mcp CLAUDE.md is an instructions file published in the GitHub repository maxkuminov/obsidian-mcp (18 stars, last pushed yesterday), licensed MIT. It adds 13,411 tokens to every session, about $0.0536 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Memory and contextResearch
Related

obsidian-mcp CLAUDE.md compared with instructions from other repositories

obsidian-mcp CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
obsidian-mcp CLAUDE.mdmaxkuminov/obsidian-mcpInstructions file18yesterdayMITA13,411Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode