crypto-link-check
This MCP server provides an offline crypto-link phishing checker plus information for an optional paid hosted full check.
Use
check_crypto_linkto check one URL or bare domain for lookalike characters, typosquats, punycode, free hosting, and bait words, returning a verdict, parsed host, matched/impersonated brand, and findings.Use its verdicts (
official,no_red_flags,caution,suspicious,dangerous) before opening links, connecting wallets, signing messages, or approving tokens.Use
full_report_infoto get the endpoint, price, and payment method for the paid x402 hosted check covering domain age, DNS, redirects, and page scanning for seed-phrase prompts or drainer code; 1–10 URLs for 0.005 USDC on Base.It never asks for keys or seed phrases; the local check makes no network request and does not visit the page, and it only judges hostname/scheme against a curated list of about 60 brands, so
no_red_flagsis not proof of safety.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@crypto-link-check check if https://phantorn.app is a fake Phantom wallet site"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Crypto Link Check
Is that MetaMask / Phantom / Ledger / exchange link real? Check it before you (or your agent) connect a wallet or sign anything.
October 2026 alone brought fake Phantom, Jupiter and Bitget browser extensions, a hijacked NGRAVE push alert that sent users to a seed-phrase page, and exchange phishing mails. AI agents can now trade from exchange accounts and pay for things on their own, so they click links too. This tool gives them (and you) a quick, read-only second opinion.
Free, offline: CLI and MCP server that compare a link with the official domains of ~60 wallets, exchanges and dapps and flag lookalike characters (
0/o,rn/m, Cyrillic), typosquats, punycode, free hosting and bait words like "claim" or "sync".Free API + remote MCP:
https://crypto-link-check.nextill-pos.workers.devPaid full check (optional, 0.005 USDC via x402 on Base): adds domain registration age (RDAP), DNS, redirect tracing and one fetch of the page scanned for seed-phrase prompts and token-approval drainer code. 1-10 URLs per call.
It never asks for keys or seed phrases. Verdicts are heuristics, not guarantees.
CLI
npx devx-link-check phantorn.app metamask.io
# verdict: dangerous phantorn.app
# [danger] phantorn.app spells the official phantom.app with lookalike characters ...
# verdict: official metamask.ioExit code: 0 official / no red flags, 2 caution / suspicious / dangerous, 1 error. --json for machine output.
Related MCP server: onchain-forensics
MCP
Local (stdio), runs offline:
{ "mcpServers": { "crypto-link-check": { "command": "npx", "args": ["-y", "devx-link-check", "mcp"] } } }Remote (streamable HTTP, no key): https://crypto-link-check.nextill-pos.workers.dev/mcp
Tools: check_crypto_link(url) and full_report_info().
HTTP API
Route | Price | What |
| free | lexical verdict + findings |
| 0.005 USDC (x402 v2 | full check, 1-10 URLs |
| free | the official-domain list we use |
| free | discovery |
The paid route answers 402 with a PAYMENT-REQUIRED header; any x402 client can pay. Settlement happens only after the report is produced. Humans can pay from a browser wallet on the web page. Terms: /terms.
Verdicts: official, no_red_flags, caution, suspicious, dangerous. Same-name-different-TLD domains are only caution because the brand may own them.
Official-domain list
cli/src/brands.js. It is curated by hand and incomplete. PRs that add brands or official domains (with a source) are welcome.
Layout
cli/: npm packagedevx-link-check(CLI + stdio MCP)worker/: the Cloudflare Worker behind the hosted API
MIT. Made by DevX. Contact madebydevx@gmail.com.
Guide
Related free tool: 7702 Sweeper Check — is a sweeper on your wallet?
Related free tool: Address Poisoning Check — is the address you are about to pay a poisoned lookalike?
Available Tools
2 toolscheck_crypto_linkCheck a crypto link for phishingARead-onlyIdempotentInspect
Checks one URL or domain for signs of crypto phishing and tells you whether it belongs to a known wallet, exchange or dapp. Use it before an agent or user opens a link to connect a wallet, sign a message or approve a token, for example a link from an email, DM, ad, search result or support chat. It runs locally and offline. It makes no network request, does not visit the page, and does not ask for keys. It compares the hostname with a hand-curated list of official domains for about 60 brands and flags lookalike characters (0/o, rn/m, Cyrillic), typosquats, punycode, free hosting and bait words such as 'claim', 'airdrop' or 'sync'. Returns: verdict (official = on the official list; no_red_flags = nothing found, which is not proof of safety; caution = one warning, such as the brand name on a different TLD; suspicious = two or more warnings; dangerous = impersonation or lookalike), the parsed host and its Unicode form, the official brand it matches or the brand it seems to imitate, and a list of findings. Each finding has a level, code and message. It does not check domain age, DNS, redirects or page content. For those, see full_report_info (a paid hosted check). Unknown brands are not covered.
Inputs: url (required). Returns: url, host, input, verdict, +7 more
full_report_infoHow to buy the full hosted link checkARead-onlyIdempotentInspect
Returns the endpoint, price and payment method of the optional paid hosted check. It does not call the API and does not pay or sign anything. The paid check covers what check_crypto_link cannot see offline: domain registration age (RDAP), DNS resolution, the redirect chain, and one fetch of the page scanned for seed-phrase prompts and token-approval drainer code. It accepts 1-10 URLs per call for 0.005 USDC on Base via x402. Use it when check_crypto_link returns no_red_flags or caution for a link that someone is about to trust with a wallet and you want more evidence. If check_crypto_link already says dangerous or official, you usually do not need it. Returns: method, url_template, price_usdc, network, asset, protocol, max_urls, settles and docs.
Inputs: no named parameters. Returns: docs, asset, method, network, +5 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools are mostly distinct: one performs a local link check and the other only returns metadata about a paid hosted service. However, the name 'full_report_info' could be mistaken for an actual deep-scan operation, and only the description prevents that misselection.
Both names use snake_case and are readable, but they follow different structural conventions: an action verb phrase (check_crypto_link) versus a noun-ish descriptor (full_report_info). Minor deviation rather than inconsistency.
Two tools is thin for a security-checking server; a single checking tool plus a metadata/info tool is borderline. The narrow scope keeps it defensible, but it feels under-scoped.
There is a notable dead end: the server advertises a paid hosted check but exposes no tool to actually invoke it, only an info endpoint. There is also no batch-check tool or way to list supported brands, so agents cannot complete the advertised deeper-verification workflow.
Maintenance
Related MCP Connectors
Safety checks before you sign, pay or install: contract, token and package scans, phishing, OFAC.
Read-only crypto safety: token honeypot checks, EIP-712 signature decode, approval scans.
Free crypto scam checks: token and contract risk, scam or sanctioned addresses, phishing links.
Crypto transaction safety tool — screens addresses and transactions for scam risk.
Related MCP Servers
- AlicenseAqualityDmaintenanceNon-custodial agent wallet with a transaction preflight that decodes an unsigned EVM tx and flags drain patterns (unlimited/large approval, approve-all, token & NFT transferFrom, proxy upgrade, on-chain permit, approvals hidden in multicall) before signing.91MIT
- AlicenseBqualityDmaintenanceProvides on-chain forensic checks for evaluating transaction risks, including token verification, rug-pull detection, and fund tracing, using public blockchain endpoints.1239 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables MCP-compatible AI agents to scan crypto transactions for scam addresses, clipboard-hijack patterns, and typos, as well as report scam addresses and check usage, directly in conversation.-
- AlicenseNot gradedqualityCmaintenanceEnables checking suspicious URLs for risk without opening them, providing full checkups and lookalike detection.MIT