vault-kv-mcp
A read-only MCP server that provides tools to read, list, and inspect secrets from HashiCorp Vault's KV secrets engine (versions 1 and 2) using a Vault token.
README
vault-kv-mcp
A read-only MCP (Model Context Protocol) server that exposes the HashiCorp Vault KV secrets engine (versions 1 and 2) as tools. Every tool performs only non-destructive reads — there are no write, delete, or destroy operations. It authenticates with a Vault token and speaks MCP over stdio, so it works with local MCP clients such as Claude Desktop.
Tools (all read-only)
| Tool | Description |
|---|---|
vault_kv_read |
Read a secret (latest or a specific KV v2 version). |
vault_kv_list |
List keys / sub-folders under a path. |
vault_kv_read_metadata |
KV v2: read version history and metadata. |
vault_list_kv_mounts |
Discover available secret mounts and their KV versions. |
vault_health |
Check Vault server health / connectivity. |
KV v1 vs v2 is auto-detected per mount when kv_version is not supplied (falling back to v2).
For defense in depth, pair this with a Vault token whose policies grant only read/list capabilities on the relevant paths.
Configuration
Set these environment variables (see .env.example):
VAULT_ADDR— Vault base URL (defaulthttp://127.0.0.1:8200)VAULT_TOKEN— required Vault token, sent as theX-Vault-TokenheaderVAULT_NAMESPACE— optional, for Vault Enterprise / HCP VaultVAULT_SKIP_VERIFY— optional, settrueto skip TLS verification (dev only)
The token only needs policies granting access to the KV paths you intend to use.
Build
npm install
npm run build
Run / test with the MCP Inspector
VAULT_ADDR=http://127.0.0.1:8200 VAULT_TOKEN=hvs.xxxx npm run inspector
Use with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"vault-kv": {
"command": "node",
"args": ["/absolute/path/to/vault-mcp/dist/index.js"],
"env": {
"VAULT_ADDR": "http://127.0.0.1:8200",
"VAULT_TOKEN": "hvs.your-token-here"
}
}
}
}
Quick local Vault for testing
vault server -dev # prints a Root Token and unseal info
export VAULT_ADDR=http://127.0.0.1:8200
export VAULT_TOKEN=<root-token-from-output>
vault kv put secret/demo username=app password=s3cr3t
Then ask your MCP client to read secret/demo.
Security notes
- The server only ever reads from Vault; it never writes, deletes, or destroys secrets.
- The token is read only from the environment and never logged.
- Prefer a short-lived, least-privilege token (read/list only) over a root token.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.