ProvenPaid MCP server
com.provenpaid/mcp·v0.2.0Check an x402 endpoint before your agent pays it: challenge validity and on-chain payer evidence.
not scanned yet
0 stars
Reviews
Write oneNobody has reviewed ProvenPaid yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
ProvenPaid tools
No tool declarations could be read from the package source. They show once the server is installed.
How ProvenPaid compares
Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 82/100
- higher than 65% of 4,460 AI and agents servers
Release history: 2 versions in the registry since 2026-10-09, latest 2026-10-11, typically one every 2 days.
Install ProvenPaid in Claude Code, Cursor or VS Code
claude mcp add mcp -- uvx provenpaid-mcp
What the publisher says
From the ProvenPaid repository's README, as published. We do not edit it. Read it on GitHub
provenpaid-mcp
Give your agent a pre-payment check for paid x402 endpoints. Evidence before you pay.
<!-- mcp-name: com.provenpaid/mcp -->
ProvenPaid checks an x402 endpoint before your agent relies on it: is its payment challenge valid, and do real wallets actually pay it on-chain, come back, and how concentrated are payments? This MCP server exposes that check as a tool and pays for it automatically, about $0.01 in USDC on Base per check, from a wallet you control.
Use it once before relying on an unfamiliar endpoint, not before every call.
Tools
You are not charged when the URL is invalid or unsafe, is not an x402 endpoint, or offers no payment option in USDC on Base. Results are evidence, not a verdict: ProvenPaid does not verify service quality or that an endpoint controls its receiving address.
Install
Requires Python 3.11+. With uv:
uvx provenpaid-mcpor pip install provenpaid-mcp.
Claude Desktop / Claude Code / Cursor
{
"mcpServers": {
"provenpaid": {
"command": "uvx",
"args": ["provenpaid-mcp"],
"env": {
"PROVENPAID_WALLET_PRIVATE_KEY": "0x...",
"PROVENPAID_DAILY_LIMIT": "20"
}
}
}
}Wallet safety
- Use a dedicated wallet holding only a few dollars of USDC on Base. Never use a wallet with meaningful funds.
- The key stays on your machine, in memory only. It is never logged, returned by a tool, or sent anywhere; only signed payment authorizations leave the process.
- No ETH is needed: the payment facilitator pays network fees.
Guardrails enforced before anything is signed:
How it works
The server calls GET https://api.provenpaid.com/v1/check?url=.... The API answers HTTP 402 with x402 payment requirements; the official x402 SDK signs a single-use USDC authorization within the guardrails above and retries. Settlement happens only when the check delivers evidence. Methodology: https://api.provenpaid.com/methodology
License
MIT
Nothing above is checked by us. What we check is on the safety report.
ProvenPaid: common questions
- Is ProvenPaid MCP server safe?
- Mostly: it is graded B (82/100). Read the ProvenPaid safety report
- How do I install ProvenPaid?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does ProvenPaid need an API key?
- Yes. The registry entry asks for
PROVENPAID_WALLET_PRIVATE_KEY. - Is ProvenPaid maintained?
- The last commit was 26 days ago (2026-10-09). The latest release is v0.2.0.
- What can I use instead of ProvenPaid?
- Servers from other publishers that do the same job: HOL Guard MCP server, Decionis MCP server and Watch Skill MCP server.
Alternatives to ProvenPaid
Same job from other publishers: the closest match first, then the best rated.
- HOL GuardLocal-first AI agent security evidence and approval workflows through HOL Guard's stdio MCP server.not reviewedEstablishedA
- DecionisAuthorize consequential AI agent actions before executionnot reviewedEstablishedA
- Watch SkillWatch video and live sessions, keep timestamped evidence, and verify an agent's own work.not reviewedEstablishedB
- DCC-MCP CoreAgent control plane for DCC and engine hosts: one MCP endpoint, one tool contract.not reviewedEstablishedA
- OpenWorkProofVerifiable agent evidence; opt-in Codex guard and reference SOC alert review.not reviewedEstablishedA