vault-mcp: Instructions file for Claude Code

CLAUDE.md

vault-mcp CLAUDE.md is an instructions file for Claude Code from wakita181009/vault-mcp. It costs 1,073 tokens per session, scanned A, original, MIT.

Instructions for maintaining vault-mcp, a remote service that reads and changes a private GitHub-hosted Obsidian vault. Obsidian is a note-taking app that stores notes as files, and the instructions protect note paths, links, and restricted folders.

In plain words
What is it for?
Use them when changing vault-mcp, especially its note reading, writing, or deletion logic.
Why use it?
They prevent code changes from adding unsafe file operations or bypassing path and file-type checks.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions Claude Code.

This is wakita181009/vault-mcp's own configuration. It tells Claude Code how to work on vault-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 vault-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to wakita181009/vault-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/wakita181009/vault-mcp/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/wakita181009/vault-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 vault-mcp CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/wakita181009/vault-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/wakita181009/vault-mcp/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/wakita181009/vault-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/wakita181009/vault-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 vault-mcp CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/wakita181009/vault-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/wakita181009/vault-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,073 This file is loaded in full into every session.
When invoked 1,073 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 $0.01073 $0.01073
Opus 5.5 $0.00429 $0.00429
Sonnet 5.5 $0.00215 $0.00215
Haiku 4.5 $0.00107 $0.00107

Measured 8d ago against content hash 0eb3f88c7268, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

vault-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 8d ago.

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 · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

vault-mcp

Authenticated remote MCP server (Cloudflare Workers) that exposes a private, GitHub-hosted Obsidian vault to claude.ai and Claude Code. Reads (list/read/search) plus two note mutations: write_note (create/overwrite) and delete_note — no rename/move. The target repo and access policy are configured through secrets; path-scope overrides default in src/config.ts, so wrangler.jsonc ships generic (no vars).

Invariants — keep these intact when changing code

  • Notes-only, no rename/move. The mutations are write_note (create/overwrite a markdown note) and delete_note (remove one). Never add a rename/move tool — the GitHub API has no atomic move and it would silently break Obsidian wikilinks. Never let a mutation touch non-note files — writeNote and deleteNote in vault.ts must keep enforcing the .md/.markdown extension check.
  • Policy lives in vault.ts, never in a backend. Storage backends (src/storage/) implement VaultStorage and only move bytes. All guards run in VaultClient before a path reaches a backend, so a new backend cannot weaken them. Add backends by implementing the interface and wiring it in vaultClientFromEnv.
  • Writes and deletes obey the same path policy as reads. write_note and delete_note must run every guard read_note does — ../absolute rejection and VAULT_ALLOWED_PREFIXES / VAULT_DENIED_PREFIXES — so a mutation can never reach .git/, .obsidian/, or agent dirs. Preserve this when touching write/delete logic.
  • The login allowlist gates the whole toolset. An identity absent from the configured provider's allowlist (VAULT_ALLOWED_GITHUB_LOGINS for AUTH_PROVIDER=github, VAULT_ALLOWED_GOOGLE_EMAILS for google, VAULT_ALLOWED_OIDC_EMAILS for oidc) must get zero tools, not a reduced set — see guard.ts. Login providers live in src/auth/providers/; the flow in auth/handler.ts is provider-agnostic. Google and the generic oidc provider share createOidcProvider; an email identity must be verified (email_verified: true) before it can match. The only exception is the explicit OIDC_ALLOW_UNVERIFIED_EMAIL=true opt-in for IdPs that never send the claim — an explicit email_verified: false is rejected regardless. Keep the OIDC discovery document's issuer check in providers/oidc/discovery.ts: without it a hostile host could supply foreign endpoints.
  • VAULT_GITHUB_TOKEN stays single-repo scoped. It needs Contents: Read and Write (Write covers both write_note and delete_note), but only on the vault repo. Never widen it beyond that one repo or swap in a broader token.
  • Path safety. vault.ts enforces VAULT_ALLOWED_PREFIXES / VAULT_DENIED_PREFIXES (defaults in src/config.ts; DEFAULT_DENIED_PREFIXES must keep hiding .git/, .obsidian/, and agent dirs) and rejects .. traversal and absolute paths in read_note, write_note, and delete_note. Preserve this when touching read/write/delete logic.

Read the full file on GitHub · 59 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. 8d ago Changed · +10 lines · +227 tokens per session 0eb3f88c7268
  2. 22d ago Changed · +9 lines · +197 tokens per session 20a335ac3d1d
  3. 2mo ago First seen · 40 lines · 649 tokens per session scan A f404136513a6

Subscribe to this mod's changes

vault-mcp CLAUDE.md is an instructions file published in the GitHub repository wakita181009/vault-mcp (7 stars, last pushed 5d ago), licensed MIT. It adds 1,073 tokens to every session, about $0.0043 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 2026-08-31.

Related

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

vault-mcp CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
vault-mcp CLAUDE.mdwakita181009/vault-mcpInstructions file75d agoMITA1,073Claude 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