
Clawvault Mcp Server
io.github.andrewszkPublished by GitHub account andrewszkv0.1.1Updated Oct 25, 2026
AI agent payment security - spending limits, whitelists, and human approval.
Overview
ClawVault MCP server lets an AI assistant request crypto payments that are checked against spending limits, whitelists and human approval before they execute.
- What you can do
- Request payments in a token to a recipient address on a chosen chain, with an optional reason, subject to the vault's security rules. Pre-check whether a payment would be auto-approved or need manual approval. Look up the status of a pending payment by ID and list recent transactions with pagination. Read vault status, including wallet address, balances and current spending limits. Emergency-freeze an agent so all its payment requests are blocked.
- When to use it
- Use it when you want an assistant to initiate real crypto payments but keep them under spending limits, recipient whitelists and human approval rather than letting the agent pay freely. It suits agent-driven purchasing such as paying for API credits, with oversight from a dashboard or Telegram.
- Requirements
- A local Node.js runtime to run the npm package, either installed globally or via npx. A ClawVault account and an agent API key supplied in the CLAWVAULT_API_KEY environment variable. Network access to the ClawVault API; the base URL can be overridden with CLAWVAULT_API_URL.
Installation
In SourceWeft
- Open Clawvault Mcp Server in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
clawvault-mcp-server
MCP (Model Context Protocol) server for ClawVault - AI agent payment security layer.
Installation
Or use directly with npx (no install needed):
Usage with Claude Desktop
Add to your Claude Desktop config:
Mac: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Restart Claude Desktop after saving.
Getting Your API Key
- Go to clawvault.cc and sign up
- Navigate to the Agents page
- Create a new agent or use an existing one
- Copy the API key (format:
cv_live_...)
Available Tools
Once connected, Claude has access to these tools:
request_payment
Request a payment through ClawVault. The payment will be evaluated against your security rules.
check_limits
Check if a payment would be allowed before making it. Returns whether it would auto-approve or need manual approval.
get_payment_status
Check the status of a pending payment by ID.
list_transactions
List recent payment transactions with pagination.
get_vault
Get vault status including wallet address, balances, and current spending limits.
freeze_agent
Emergency freeze an agent to block all its payment requests.
How It Works
- You ask Claude to pay for something
- Claude uses
request_paymenttool - ClawVault checks your rules:
- Within rules → auto-approved, payment executes
- Outside rules → pending, you get a Telegram/dashboard notification
- You approve or deny from Telegram or the web dashboard
- Claude gets the result
Environment Variables
For Other AI Agents
Not using Claude Desktop? Any AI agent can use ClawVault by reading the skill documentation:
This returns plain text that teaches the agent how to use the ClawVault API.
Development
Links
- Website: clawvault.cc
- Docs: clawvault.cc/docs
- API: api.clawvault.cc
- Skill docs: api.clawvault.cc/skill
License
MIT
Source: README.md at commit d3c5e7b
Tools
0Version history
1- v0.1.1LatestSep 16, 2026
Score
40% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.