Skip to content

Money Maker x402 tools: EVM reads + AI tasks MCP server

dev.workers.moneymaker-ai.ai-policy-kit/x402-toolsai-policy-kit.moneymaker-ai.workers.devstreamable-httpv1.1.0Scanned

x402 pay-per-call: EVM balances, tx decode, transfers; AI extract, classify, embed, speech, image

28 tools: 3 do something that can't be undone, 20 change things and 5 only read.

Tools: 0 delete, 3 irreversible, 20 write, 5 read.
  • Reads private data · 0
  • Reads text others write · 27
  • Can send data out · 3
  • 4 findings

Tools listed without sign-in. Remote posture

With Anlyon

3 tools here do something that can't be undone. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Money Maker x402 tools: EVM reads + AI tasks MCP server can do

The Money Maker x402 tools: EVM reads + AI tasks MCP server is published by ai-policy-kit.moneymaker-ai.workers.dev, a domain the MCP Registry verified. It has 28 tools. It can do things that cannot be undone, such as sending, paying or publishing (usdc_payment, token_transfers and address_transfers) and create or change data (20 tools: repo_ai_policy, x402_lint, erc8004_agent, gas_now and 16 more), but cannot delete data. 5 tools only read.

27 tools bring in text other people can write (repo_ai_policy, x402_lint, usdc_payment and 24 more); 3 tools can send data out of your system (x402_lint, usdc_payment and web_extract).

The fixed checks matched 4 findings (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 8, 2026, version 1.1.0.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

28 tools

Can't be undone 3Claude Code: ask
  • address_transfersRecent token activity for an address on one chain: ERC-20 Transfer logs where it is sender or recipient, for the chain's major tokens (USDC, USDT, WETH, BTC wrappers, DAI, chain to…Reads text others write1 finding

    Why it is in this group

    • The name "address_transfers" says "transfer": a word for a change that another call does not undo.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "address_transfers" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "address_transfers" says "transfer": a word for a change that another call does not undo.

      What this check looks for
    Description, as the server wrote it
    Recent token activity for an address on one chain: ERC-20 Transfer logs where it is sender or recipient, for the chain's major tokens (USDC, USDT, WETH, BTC wrappers, DAI, chain token) or up to 10 tokens you pass, within the last N blocks (max 10,000, the free public-RPC log limit: ~5.5 h on base/optimism/polygon, ~33 h on ethereum, ~75 min on bsc, ~40 min on arbitrum). Newest first, max 50, with direction, symbol, decimal amount and block time. Native-coin transfers and older history are not included. If every free RPC is rate-limited you get 502 and are not charged. Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/address-transfers.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • token_transfersPayment proof for any EVM tx: success, block, UTC time, confirmations, sender, and every ERC-20 Transfer (token, symbol, decimals, from, to, exact raw + decimal amount) plus ERC-72…Reads text others write1 finding

    Why it is in this group

    • The name "token_transfers" says "transfer": a word for a change that another call does not undo.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "token_transfers" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "token_transfers" says "transfer": a word for a change that another call does not undo.

      What this check looks for
    Description, as the server wrote it
    Payment proof for any EVM tx: success, block, UTC time, confirmations, sender, and every ERC-20 Transfer (token, symbol, decimals, from, to, exact raw + decimal amount) plus ERC-721 transfers, decoded from the receipt. Chains: base, ethereum, bsc, arbitrum, optimism, polygon, gnosis, celo. Price: $0.005 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/token-transfers.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • usdc_paymentDid this transaction pay USDC on Base to an address?Reads text others writeCan send data out1 finding

    Why it is in this group

    • The name "usdc_payment" says "payment": a word that usually means money moves, which cannot be taken back.
    • The input addresses people (input.to): what it sends cannot be taken back.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a destination (input.to): what the tool writes goes there.

    Findings

    • Warn

      The server marks "usdc_payment" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "usdc_payment" says "payment": a word that usually means money moves, which cannot be taken back.

      What this check looks for
    Description, as the server wrote it
    Did this transaction pay USDC on Base to an address? Returns success, confirmations, every USDC transfer in the tx, the total received by `to`, and whether it meets `min`. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/usdc-payment.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
Changes things 20Claude Code: ask
  • address_infoWhat is this address on each chain: unused EOA, active EOA, EIP-7702-delegated EOA (with delegate), ERC-20, ERC-721, ERC-1155, Safe multisig (threshold, owners, version), or other …Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    What is this address on each chain: unused EOA, active EOA, EIP-7702-delegated EOA (with delegate), ERC-20, ERC-721, ERC-1155, Safe multisig (threshold, owners, version), or other contract; EIP-1967/beacon/zos/EIP-1167 proxy with implementation and admin; owner(); nonce, native balance and code size. From bytecode, storage slots and probe calls via free public RPCs. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/address-info.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • ai_classifyZero-shot classification task: send text (max 8,000 chars) and 2-50 labels; get one label (or every applicable label with multi_label=true) guaranteed to be from your list, plus a …Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Zero-shot classification task: send text (max 8,000 chars) and 2-50 labels; get one label (or every applicable label with multi_label=true) guaranteed to be from your list, plus a one-sentence reason. Invalid output → 422, not charged. Runs on our Cloudflare Worker using Workers AI (IBM Granite 4.0 micro, Apache-2.0). No calibrated probabilities. Limited daily capacity: 503 (not charged) when used up. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: POST /api/ai/classify.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: false
  • ai_embedEmbed 1-32 texts (each up to 4,000 chars) with BAAI bge-m3 (multilingual, MIT) on our Cloudflare Worker via Workers AI.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Embed 1-32 texts (each up to 4,000 chars) with BAAI bge-m3 (multilingual, MIT) on our Cloudflare Worker via Workers AI. Add query= to also get the texts ranked by cosine similarity (a cheap semantic search / rerank step); include_vectors=false returns only the ranking. Limited daily capacity: 503 (not charged) when used up. Price: $0.0005 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: POST /api/ai/embed.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: false
  • ai_extractStructured extraction task: send text (max 12,000 chars) and a JSON Schema; get back JSON that has been validated against your schema (type, properties, required, enum, items, min/…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Structured extraction task: send text (max 12,000 chars) and a JSON Schema; get back JSON that has been validated against your schema (type, properties, required, enum, items, min/max, pattern, anyOf/oneOf). One automatic repair attempt; if it still does not validate you get 422 and are not charged. Fixed extraction prompt (copy, don't invent). Runs on our Cloudflare Worker using Workers AI (IBM Granite 4.0 micro, Apache-2.0). Not a chat or model-access API. Limited daily capacity: 503 (not charged) when used up. Price: $0.004 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: POST /api/ai/extract.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: false
  • ai_imageGenerate one 1024×1024 image from a short prompt (max 600 chars) with FLUX.1 [schnell] (Apache-2.0) on our Cloudflare Worker via Workers AI.Reads text others write1 finding

    Why it is in this group

    • The description's first verb is "generate": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "ai_image" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "generate": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Generate one 1024×1024 image from a short prompt (max 600 chars) with FLUX.1 [schnell] (Apache-2.0) on our Cloudflare Worker via Workers AI. Before generation we detect the medium you asked for (photo, illustration, flat vector, logo/icon, watercolor, oil painting, pencil sketch, ink, anime, comic, pixel art, 3D render, digital painting) and rewrite the prompt to honour it, with style-specific quality guidance (e.g. real lens and natural light for photos, clean 2D linework for illustrations); no medium named = natural photo. The medium is never switched and nothing unrequested is added; set style to force one, or enhance=false to use your prompt as written. Both your prompt and the final prompt go through our content screen (keyword rules + an LLM check): sexual content, minors, violence/weapons, real or identifiable people, hate and illegal content are refused with 400 and you are not charged. The image itself is not classified. Returns a base64 JPEG plus the final prompt and detected style. Capped at a small number of images per day: 503 (not charged) when used up. Price: $0.01 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: POST /api/ai/image.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: false
  • ai_transcribeTranscribe one audio file (mp3/wav/m4a/ogg/webm, max 1 MB, via public https audio_url or audio_base64) with OpenAI Whisper large-v3-turbo (MIT) on our Cloudflare Worker via Workers…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Transcribe one audio file (mp3/wav/m4a/ogg/webm, max 1 MB, via public https audio_url or audio_base64) with OpenAI Whisper large-v3-turbo (MIT) on our Cloudflare Worker via Workers AI, with voice-activity filtering. Returns text, language, duration and optional segments + WebVTT. Audio is not stored. Limited daily capacity: 503 (not charged) when used up. Price: $0.005 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: POST /api/ai/transcribe.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: false
  • block_at_timeThe last block at or before a given time (unix seconds/ms or ISO 8601) on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo, with its timestamp and the next block's,…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    The last block at or before a given time (unix seconds/ms or ISO 8601) on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo, with its timestamp and the next block's, so you can pin eth_call reads, audits and snapshots to a moment. Batched k-ary search over live block headers (about 7 round trips). Very old history may be unavailable on public nodes. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/block-at-time.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • domain_checkOne call for a domain (or URL): A/AAAA, NS, SOA, MX (incl.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    One call for a domain (or URL): A/AAAA, NS, SOA, MX (incl. null MX), TXT, CAA, DNSSEC validation flag, parsed SPF (all-qualifier, includes, lookup estimate, verdict), DMARC (policy, sp, pct, rua), MTA-STS presence, site-verification tokens, and one HTTPS GET of / (status, final URL, HSTS, server), plus a plain-language issues list. Read live via DNS-over-HTTPS (Cloudflare 1.1.1.1, Google fallback). No WHOIS. Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/domain-check.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • email_checkIs this address worth sending to?Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Is this address worth sending to? Syntax check, then the domain's mail route via DNS-over-HTTPS: domain exists, MX hosts (or A fallback), null MX (domain refuses mail), SPF and DMARC presence, role account (info@, support@…) and plus-tag flags, and a verdict. It does not contact the mail server, so it cannot confirm that the mailbox itself exists. Price: $0.001 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/email-check.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • erc20_allowanceHow much can each spender pull from an owner's tokens?Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    How much can each spender pull from an owner's tokens? For a token on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo: allowance(owner, spender) for up to 10 spenders, owner balance, symbol, decimals, whether each spender is a contract, EOA or 7702-delegated EOA, and flags for unlimited approvals and approvals to plain EOAs. Use before or after approving, or to audit what to revoke. Price: $0.001 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/erc20-allowance.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • erc8004_agentKnow-your-agent check before you pay or hire an agent: reads the ERC-8004 Identity Registry (0x8004A169…a432) at a pinned block on base, ethereum, bsc, arbitrum, optimism, polygon,…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Know-your-agent check before you pay or hire an agent: reads the ERC-8004 Identity Registry (0x8004A169…a432) at a pinned block on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo and returns ownerOf, getAgentWallet and tokenURI, fetches the registration file (https/ipfs/data) with its sha256, and checks that the file's agentWallet and registrations[] match the chain. Includes the exact eth_call data so anyone can re-check. Price: $0.01 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/erc8004-agent.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • gas_nowCurrent gas price, base fee, priority fee, block fullness and the cost of a native transfer (21k gas) and an ERC-20 transfer (65k gas) in the native coin, for base, ethereum, bsc, …Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Current gas price, base fee, priority fee, block fullness and the cost of a native transfer (21k gas) and an ERC-20 transfer (65k gas) in the native coin, for base, ethereum, bsc, arbitrum, optimism, polygon, gnosis and celo (or one chain with ?chain=). Price: $0.001 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/gas-now.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • policyAI_POLICY.md (strict | disclose-and-own | agent-friendly) for your project plus a PR template with an AI-disclosure section, as JSON.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    AI_POLICY.md (strict | disclose-and-own | agent-friendly) for your project plus a PR template with an AI-disclosure section, as JSON. Same text as the free web generator on this site. Price: $0.01 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/policy.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • repo_ai_policyBefore a coding agent opens a PR: does this GitHub repo ban, restrict or allow AI-assisted contributions?Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Before a coding agent opens a PR: does this GitHub repo ban, restrict or allow AI-assisted contributions? Reads AI_POLICY/CONTRIBUTING/AGENTS.md/PR template/SECURITY and returns a verdict, evidence lines with file:line, and build/test commands from AGENTS.md. Keyword heuristic: read the evidence. Price: $0.01 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/repo-ai-policy.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • token_balancesTyped balances for one address: native coin plus USDC, USDT, WETH, BTC wrappers, DAI and the chain token on base, ethereum, arbitrum, optimism, polygon and bsc (gnosis/celo on requ…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Typed balances for one address: native coin plus USDC, USDT, WETH, BTC wrappers, DAI and the chain token on base, ethereum, arbitrum, optimism, polygon and bsc (gnosis/celo on request), read with one Multicall3 eth_call per chain at a stated block. Or pass chain= and tokens= (up to 20 addresses) for your own list; symbol/decimals are read on-chain. Returns raw and decimal amounts plus a nonzero summary. No USD prices. Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/token-balances.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • token_infoFor a token contract on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo: name, symbol, decimals, total supply (raw and decimal), code size, owner() if any, and whe…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    For a token contract on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo: name, symbol, decimals, total supply (raw and decimal), code size, owner() if any, and whether it is an upgradeable proxy (EIP-1967 or legacy OpenZeppelin slots) with implementation and admin addresses. Price: $0.001 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/token-info.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • token_riskIs this token centrally controllable?Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Is this token centrally controllable? Deterministic screen for a token on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo, from bytecode and eth_call only (no third-party risk API): upgradeable proxy and admin, owner and whether renounced, privileged functions found by selector scan of the token and its implementation (mint, pause, blacklist, tax/fee setters, max-tx limits, trading switch, fee exclusion, rescue), owner/self/burn share of supply, a 0-100 score with one reason per flag, and a list of what was not checked (no DEX sell simulation). Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/token-risk.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • tx_decodeTyped decode of any tx on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo (chain optional: all 8 are searched): success/reverted, method signature from a known-sel…Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Typed decode of any tx on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis or celo (chain optional: all 8 are searched): success/reverted, method signature from a known-selector table (args for transfer/approve/transferFrom), every log decoded where known (ERC-20/721/1155 transfers with symbols and decimal amounts, approvals with an unlimited flag, WETH deposit/withdraw, Uniswap V2/V3/V4 swaps, ownership/upgrade, EIP-3009), the sender's net token flows, fee incl. L1 fee, and for reverted txs a decoded revert reason (approximate replay at the previous block). Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/tx-decode.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • x402_lintProbe any public x402 endpoint and validate its 402 challenge: version, accepts[], CAIP-2 network, atomic amounts, payTo checksum, native-USDC asset, EIP-712 extra, timeout, Bazaar…Reads text others writeCan send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Probe any public x402 endpoint and validate its 402 challenge: version, accepts[], CAIP-2 network, atomic amounts, payTo checksum, native-USDC asset, EIP-712 extra, timeout, Bazaar discovery info, and /openapi.json presence. Returns errors, warnings and human-readable prices. Price: $0.005 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/x402-lint.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
Only reads 5Show 5 read-only tools
  • ai_statusFree.No findings

    Why it is in this group

    • The name "ai_status" says "status": a word for reading.
    Description, as the server wrote it
    Free. Today's remaining free Workers AI capacity for the ai_* tools (they answer 'capacity used up' before any payment when the daily cap is reached).

    Annotations the server declares

    • readOnlyHint: true
  • ens_resolveResolve an ENS name or Basename (alice.eth, alice.base.eth) to its EVM and Base address records, with resolver, url/avatar/com.twitter/description text records and whether the addr…Reads text others writeNo findings

    Why it is in this group

    • The name "ens_resolve" says "resolve": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe description's first sentence says "resolve" and "url": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Resolve an ENS name or Basename (alice.eth, alice.base.eth) to its EVM and Base address records, with resolver, url/avatar/com.twitter/description text records and whether the address's primary name points back (reverse_matches). Or pass address= for its forward-verified primary name (mainnet and Base/ENSIP-19). Read live via the ENS Universal Resolver with CCIP-read. Price: $0.001 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/ens-resolve.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • tx_findGot a tx hash but not the chain?Reads text others writeNo findings

    Why it is in this group

    • The name "tx_find" says "find": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Got a tx hash but not the chain? Searches base, ethereum, bsc, arbitrum, optimism, polygon, gnosis and celo in parallel and returns where it was mined: status (success/reverted/pending), block, UTC time, confirmations, from, to, native value, method selector, nonce, gas used, effective gas price, fee (incl. L1 fee when the receipt has one), log and ERC-20 transfer-log counts, explorer link. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/tx-find.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • wallet_snapshotOne call, 8 chains: native balance, USDC/USDT balances (main contracts per chain, addresses returned), nonce (txs sent), and account type (EOA, EIP-7702-delegated EOA with its dele…Reads text others writeNo findings

    Why it is in this group

    • The name "wallet_snapshot" says "snapshot": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    One call, 8 chains: native balance, USDC/USDT balances (main contracts per chain, addresses returned), nonce (txs sent), and account type (EOA, EIP-7702-delegated EOA with its delegate, or contract) on base, ethereum, bsc, arbitrum, optimism, polygon, gnosis and celo, plus USDC/USDT totals and the chains where the address is active. Price: $0.003 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/wallet-snapshot.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true
  • web_extractFetch one public web page politely (checks robots.txt first and refuses disallowed paths; identifies itself as MoneyMakerBot) and return status, final URL after redirects, title, l…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The description's first verb is "fetch": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe name "web_extract" says "web": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "fetch" and "web": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Fetch one public web page politely (checks robots.txt first and refuses disallowed paths; identifies itself as MoneyMakerBot) and return status, final URL after redirects, title, language, meta description, canonical, favicon, Open Graph/Twitter tags, JSON-LD types, h1-h3 outline, link counts with a sample, and the readable text of <main>/<article>/<body> (scripts and styles stripped, up to max_chars, default 20,000). No JavaScript rendering. Private hosts are refused. Failed fetches return an error status, so x402 clients are not charged. Price: $0.002 USDC on Base per call, paid with x402 (MCP transport: an unpaid call returns the x402 PaymentRequired object; retry with params._meta["x402/payment"]). Same tool over HTTP: GET /api/web-extract.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
    • idempotentHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write27
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:53 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
23
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 23 tools that change things (3 irreversible, 20 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://ai-policy-kit.moneymaker-ai.workers.dev/mcp: tools/list returned 28 tools to an anonymous client, including usdc_payment (irreversible), token_transfers (irreversible), address_transfers (irreversible), repo_ai_policy (write), x402_lint (write) and 18 more

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

4 warn
  • The server marks "usdc_payment" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "usdc_payment" says "payment": a word that usually means money moves, which cannot be taken back.

    What this check looks for
  • The server marks "token_transfers" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "token_transfers" says "transfer": a word for a change that another call does not undo.

    What this check looks for
  • The server marks "address_transfers" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":true,"openWorldHint":true}; The name "address_transfers" says "transfer": a word for a change that another call does not undo.

    What this check looks for
  • The server marks "ai_image" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "generate": a word for a change.

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 1.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http x402-tools https://ai-policy-kit.moneymaker-ai.workers.dev/mcp
claude mcp add --transport http x402-tools https://ai-policy-kit.moneymaker-ai.workers.dev/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__x402-tools__tx_find",
      "mcp__x402-tools__wallet_snapshot",
      "mcp__x402-tools__ens_resolve",
      "mcp__x402-tools__ai_status"
    ],
    "ask": [
      "mcp__x402-tools__repo_ai_policy",
      "mcp__x402-tools__x402_lint",
      "mcp__x402-tools__usdc_payment",
      "mcp__x402-tools__erc8004_agent",
      "mcp__x402-tools__token_transfers",
      "mcp__x402-tools__gas_now",
      "mcp__x402-tools__token_info",
      "mcp__x402-tools__block_at_time",
      "mcp__x402-tools__erc20_allowance",
      "mcp__x402-tools__token_risk",
      "mcp__x402-tools__chainlink_price",
      "mcp__x402-tools__web_extract",
      "mcp__x402-tools__domain_check",
      "mcp__x402-tools__email_check",
      "mcp__x402-tools__token_balances",
      "mcp__x402-tools__address_info",
      "mcp__x402-tools__tx_decode",
      "mcp__x402-tools__address_transfers",
      "mcp__x402-tools__ai_extract",
      "mcp__x402-tools__ai_classify",
      "mcp__x402-tools__ai_embed",
      "mcp__x402-tools__ai_transcribe",
      "mcp__x402-tools__ai_image",
      "mcp__x402-tools__policy"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as x402-tools. Review it before you use it.

Capability tags behind 27 rules
  • mcp__x402-tools__repo_ai_policy untrusted-input
  • mcp__x402-tools__x402_lint untrusted-input, outbound
  • mcp__x402-tools__usdc_payment untrusted-input, outbound
  • mcp__x402-tools__erc8004_agent untrusted-input
  • mcp__x402-tools__token_transfers untrusted-input
  • mcp__x402-tools__tx_find untrusted-input
  • mcp__x402-tools__wallet_snapshot untrusted-input
  • mcp__x402-tools__gas_now untrusted-input
  • mcp__x402-tools__token_info untrusted-input
  • mcp__x402-tools__ens_resolve untrusted-input
  • mcp__x402-tools__block_at_time untrusted-input
  • mcp__x402-tools__erc20_allowance untrusted-input
  • mcp__x402-tools__token_risk untrusted-input
  • mcp__x402-tools__chainlink_price untrusted-input
  • mcp__x402-tools__web_extract untrusted-input, outbound
  • mcp__x402-tools__domain_check untrusted-input
  • mcp__x402-tools__email_check untrusted-input
  • mcp__x402-tools__token_balances untrusted-input
  • mcp__x402-tools__address_info untrusted-input
  • mcp__x402-tools__tx_decode untrusted-input
  • mcp__x402-tools__address_transfers untrusted-input
  • mcp__x402-tools__ai_extract untrusted-input
  • mcp__x402-tools__ai_classify untrusted-input
  • mcp__x402-tools__ai_embed untrusted-input
  • mcp__x402-tools__ai_transcribe untrusted-input
  • mcp__x402-tools__ai_image untrusted-input
  • mcp__x402-tools__policy untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/dev.workers.moneymaker-ai.ai-policy-kit/x402-tools/verify?hash=7d81a09b6f964498f9759d1a158ddc82135be60c8417b2e8a8633eef34bb7ddc" | jq -e '.changed == false'

Pinned hash 7d81a09b6f96, from Nov 8, 2026, 14:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/dev.workers.moneymaker-ai.ai-policy-kit/x402-tools/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for dev.workers.moneymaker-ai.ai-policy-kit/x402-tools
[![Anlyon Radar](https://radar.anlyon.com/badge/dev.workers.moneymaker-ai.ai-policy-kit/x402-tools.svg)](https://radar.anlyon.com/servers/dev.workers.moneymaker-ai.ai-policy-kit/x402-tools)

Source: MCP Registry. Synced Oct 22, 2026, 10:17 UTC.

Money Maker x402 tools: EVM reads + AI tasks MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Money Maker x402 tools: EVM reads + AI tasks MCP server safe?

No scan can promise that. What its definitions show: 28 tools, 23 of which can change, send or delete something, and 4 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Money Maker x402 tools: EVM reads + AI tasks MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the Money Maker x402 tools: EVM reads + AI tasks MCP server do things that cannot be undone?

Yes. 3 tools can send, pay, publish or merge: usdc_payment, token_transfers and address_transfers.

What tools does the Money Maker x402 tools: EVM reads + AI tasks MCP server have?

28 tools: 3 irreversible, 20 write, 5 read. usdc_payment, token_transfers, address_transfers, repo_ai_policy, x402_lint, erc8004_agent, gas_now, token_info and 20 more.

Does the Money Maker x402 tools: EVM reads + AI tasks MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Money Maker x402 tools: EVM reads + AI tasks MCP server send data to outside services?

3 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): x402_lint, usdc_payment and web_extract.

Has the Money Maker x402 tools: EVM reads + AI tasks MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 8, 2026.