
Kaal Agent Safety (x402 paid MCP)
io.github.immoney1Published by GitHub account immoney1v1.0.0Updated Oct 29, 2026
Paid MCP tools for URL payment-safety scanning and provider recommendations; USDC on Base via x402.
Overview
Remote MCP server offering two pay-per-call tools: a URL payment-safety scan and a ranked provider directory, paid in USDC on Base via x402.
- What you can do
- Scan a target URL for payment-safety and phishing signals, returning AVOID, CAUTION or CAUTION_NONE_DETECTED plus a signed receipt
- Look up ranked providers and cost data for a category such as search, onchain data, agent safety, paid discovery or transaction execution
- Verify a returned receipt through a free replayable verification endpoint
- Discover the server's MCP and payment manifests through free well-known endpoints
- When to use it
- Useful when an assistant needs a quick automated check on whether a URL looks safe to pay or transact with, or a ranked cost comparison of providers in a category, and the caller is set up to pay small per-call fees in USDC on Base.
- Requirements
- A remote streamable-HTTP MCP endpoint; no API keys or accounts are declared. Each tools/call must carry an X-PAYMENT header with a base64 x402 exact-scheme payload, so the client needs a funded USDC-on-Base wallet and x402 payment support. Network access to the endpoint and its payment facilitator is required.
Installation
In SourceWeft
- Open Kaal Agent Safety (x402 paid MCP) in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/io.github.immoney1%2Fkaal-agent-safety/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"kaal-agent-safety-x402-paid": {
"type": "http",
"url": "https://places-wayne-experiment-facial.trycloudflare.com/mcp"
}
}
}README
kaal-agent-safety — x402 paid MCP server
Two paid MCP tools plus three free HTTP endpoints. Buyers pay USDC on Base
(per-call, x402 exact scheme, EIP-3009 transferWithAuthorization) — no API
keys, no subscriptions. Free initialize + tools/list; payment only on
tools/call (HTTP 402 with payment requirements).
Endpoints
MCP tools
safety_check(target)— $0.002 per call: flagscredentials_in_url,no_hsts,ip_literal, lookalike hosts, missing/.well-known/x402manifest; returnsAVOID / CAUTION / CAUTION_NONE_DETECTED+ signed receipt.recommend(category)— $0.005 per call: ranked whitespace/cost directory across search, onchain-data, agent-safety, paid-discovery, tx-execution.
Connect
Free: initialize, notifications/initialized, tools/list.
Paid: tools/call requires X-PAYMENT header (base64 exact-scheme payload);
server settles through an x402 facilitator and returns X-Payment-Response
with the on-chain transaction.
Run
RECEIPT_SECRET (or WALLET_FILE=<path-to-wallet-json>) signs receipts.
NETWORK is base (mainnet) or base-sepolia. Settlement is non-custodial:
funds go straight to the configured payTo wallet on every settlement.
Verified
- Local E2E: unpaid 402 → EIP-3009 signed payment → verify → settle → on-chain receipt status 1.
- Public E2E over a tunnel: paid
GET /v1/safety/checksettled with tx0xd7bd01780b75d106fb3e286b6a3c55a78a6a486ee2cf439f4d2b3d8b5b1c28d7(Base mainnet, block 53162843, gas 83208). - Revenue ledger + treasury sweep + systemd auto-restart are part of the production deployment.
License
MIT
Source: README.md at commit c74d51e
Tools
0Version history
1- v1.0.0LatestOct 29, 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 daysRequired25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declaredRequired10 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.