_ registry / mcp http-sse · checked 2h ago

x402-safety-mcp

https://x402-safety-mcp.dgdb.workers.dev

Registry code: 3f54dc911f63b33b

api record

Paid MCP tools for AI agents on Base. Pre-transaction safety: decode what you're signing (EIP-712/Permit/calldata), detect address-poisoning look-alike recipients, audit dangerous approvals, check recipients against a live on-chain malicious-address DB, or run every check in one call.

from a public catalogue that lists it, not from the operator

endpoint
https://x402-safety-mcp.dgdb.workers.dev/mcp
protocol
http-sse ·2025-06-18
authentication
none observed
public key
none — nobody has proven they own this listing · is it yours? claim it
karma
0 · newcomer
_ is it live, free and safe measured by this hub
Is x402-safety-mcp live?
Yes — it answered the hub's last check (checked 2h ago). It answered 100% of checks over the last 30 days.
Is x402-safety-mcp free to use?
No — it charges per call, from 0.25 USDC, quoted in its own 402 answers.
What tools does x402-safety-mcp have?
19 tools: recipient_check, x402_catalog_changes, preflight_guard, address_poison_check, approval_risk, threat_feed_recent, x402_reliability_feed, x402_broken_endpoints, ….
Is x402-safety-mcp safe to connect?
The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.
reachable
live
uptime, 30 days
99.6%

90 days 99.7%· all time 99.7%

latency
863ms

last good check

priced tools
18

of 19 tools

_ answered our checks, 90 days 383 checks · signed record
  • degraded → live
  • live → degraded· timeout after 20000ms
  • down → live
  • unknown → live
_ what it is for
topic
security & compliance threat & vulnerability
used for
  • decode what a transaction signs
  • check a recipient for address poisoning
  • audit token approvals
  • check x402 service reliability
takes → gives
data, text → data
tools
18 reads1 effect unclear
_ usage and payments 30 days

Calls placed through this hub's router, from its own receipts. Every caller and every payer counts the same; the chain total is counted from three payers.

accounts
0

through this hub

calls served
0

successful

paid through this hub
0 USDC

what callers paid

_ what it can do 19 tools
18 paid 1 never probed 18 of 19 classified

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

  • x402_catalog_changes reads paid never probed

    What changed in the CDP Bazaar catalog since yesterday: price changes, and payTo changes — the address you would pay MOVED, which is a pre-payment guard. Facts only (a->b), no labels.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "type": {
      "type": "string"
    },
    "limit": {
      "type": "string"
    }
  }
}
    arguments 12 lines
  • preflight_guard reads paid never probed

    One-call pre-transaction guard: bundle what you are about to sign/send and get every applicable safety check in a single payment — signature-intent decode, live threat-DB and address-poisoning check on the recipient, and a drainer audit of standing approvals. By design no combined verdict is emitted; each sub-check's flags and evidence come back verbatim so your agent decides. Replaces 4 separate paid calls with 1.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "calldata": {
      "type": "string"
    },
    "approvals": {
      "type": "array",
      "items": {
        "type": "object",
        "propertyNames": {
          "type": "string"
        },
        "additionalProperties": {}
      }
    },
    "recipient": {
      "type": "string"
    },
    "typedData": {
      "type": "object",
      "propertyNames": {
        "type": "string"
      },
      "additionalProperties": {}
    },
    "known_addresses": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  }
}
    arguments 35 lines
  • address_poison_check reads paid never probed

    Before sending funds, check if the recipient is an address-poisoning look-alike of an address you trust (shares displayed head+tail but differs in the middle). Deterministic.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "recipient"
  ],
  "properties": {
    "recipient": {
      "type": "string"
    },
    "known_addresses": {
      "type": "array",
      "items": {
        "anyOf": [
          {
            "type": "string"
          },
          {
            "type": "object",
            "required": [
              "address"
            ],
            "properties": {
              "label": {
                "type": "string"
              },
              "address": {
                "type": "string"
              }
            }
          }
        ]
      }
    }
  }
}
    arguments 36 lines
  • approval_risk reads paid never probed

    Audit standing ERC-20/Permit2 approvals for danger. Pass your current approvals ({token, spender, allowance}); we rank which to revoke — unlimited allowance, unknown spender, or the classic drainer pattern. Deterministic, no external calls.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "approvals"
  ],
  "properties": {
    "approvals": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "spender",
          "allowance"
        ],
        "properties": {
          "token": {
            "type": "string"
          },
          "spender": {
            "type": "string"
          },
          "allowance": {
            "anyOf": [
              {
                "type": "string"
              },
              {
                "type": "number"
              }
            ]
          }
        }
      }
    }
  }
}
    arguments 37 lines
  • threat_feed_recent reads paid never probed

    Live feed of freshly-detected malicious addresses on Base — address-poisoning seeders caught via zero-value transfer sprays. Poll with since=<unix> to ingest only new entries into your agent's blocklist.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "limit": {
      "type": "string"
    },
    "since": {
      "type": "string"
    }
  }
}
    arguments 12 lines
  • x402_reliability_feed reads paid never probed

    Live reliability feed of x402 endpoints on Base: which services are up, return a valid 402 envelope, and how fast — free-probed across the CDP Bazaar catalog. Filter health=healthy to route your agent only to endpoints that actually work.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "limit": {
      "type": "string"
    },
    "since": {
      "type": "string"
    },
    "health": {
      "type": "string"
    }
  }
}
    arguments 15 lines
  • x402_broken_endpoints reads paid never probed

    The x402 endpoints that are currently broken or misconfigured — unreachable, wrong status, or an invalid 402 envelope. Use it to skip services that will waste your agent's round-trips.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "limit": {
      "type": "string"
    },
    "since": {
      "type": "string"
    },
    "domains": {
      "type": "string"
    }
  }
}
    arguments 15 lines
  • x402_listing_visibility reads paid never probed

    Your Bazaar visibility card: last-30-day calls and payer percentiles against the whole catalog cohort, staleness, and health. Pass your domain to see where you actually stand.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_revenue_scope reads paid never probed

    How much does an x402 service actually earn, and where does yours sit among its peers? We measure real USDC inflows to every payTo address in the Bazaar catalog (counting only amounts that match a listed price, so ordinary transfers are excluded) and return the distribution. Pass your domain for your own revenue, calls, payers and percentile.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_listing_watch reads paid never probed

    Is an x402 service still listed in the Bazaar? Resources are removed when they go without settlement and no notice is sent. From daily catalog snapshots we report whether a domain is present today, how its resource count moved, and how long since each resource was last paid for.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_buyer_insight reads paid never probed

    Who actually pays an x402 service, and what else do those buyers purchase? Built from the payer-to-service map across the whole catalog: distinct payers, repeat rate, and the share of your buyers who also pay other x402 services. Individual buyer addresses are never disclosed.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_payment_debug paid never probed

    Would a real payment to an x402 endpoint actually be accepted? We build a signed payment at the listed price and submit it to the CDP facilitator's verify endpoint — signature, requirements and on-chain transfer simulation. Nothing settles, so no money moves. If it fails you get the facilitator's own reason. In our market-wide run, every payload-schema rejection had a resource description over 500 characters, and none of the accepted ones did.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "url": {
      "type": "string"
    },
    "domain": {
      "type": "string"
    }
  }
}
    arguments 12 lines
  • x402_sales_gap reads paid never probed

    Listed, able to charge, and still not selling? We join the Bazaar catalog with measured on-chain revenue and report the median profile of the top fifth of earners, of everyone earning anything, and of those earning nothing — next to your own. In our data the top fifth list 14 resources to the zero group's one, and write 190-character descriptions to their 370. These are correlations across services, not advice.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_ecosystem_map reads paid never probed

    Machine-readable map of the x402 service ecosystem on Base: price range, category, endpoint count, health and latency per service. Sort by resources/latency/price/newest.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "sort": {
      "type": "string"
    },
    "limit": {
      "type": "string"
    },
    "since": {
      "type": "string"
    },
    "health": {
      "type": "string"
    },
    "category": {
      "type": "string"
    }
  }
}
    arguments 21 lines
  • what_am_i_signing reads paid never probed

    Decode an EVM signature request (EIP-712 typed data, Permit/Permit2, or ERC-20 approve calldata) into a structured intent: token, spender, amount, expiry, risk_flags, action_hint. Deterministic, no LLM. Answers 'what am I about to sign?'

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "calldata": {
      "type": "string"
    },
    "typed_data": {
      "type": "object",
      "propertyNames": {
        "type": "string"
      },
      "additionalProperties": {}
    }
  }
}
    arguments 16 lines
  • x402_platform_health reads 0.25 USDC paid 2h ago

    Shared-infrastructure health across x402 services: when many endpoints sit behind the same platform, one platform incident takes them all down together. Pass domain to locate a service's platform cohort.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "domain": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • x402_market_pulse reads paid never probed

    Census of the x402 market: how many services appeared or disappeared since yesterday, price-median drift over time, and network mix. kind=new|gone for the actionable churn.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "kind": {
      "type": "string"
    },
    "limit": {
      "type": "string"
    }
  }
}
    arguments 12 lines
  • rpc_health_map reads 0.25 USDC paid 2h ago

    Which Base public RPCs are serving a stale chain tip. We ask all major public RPCs for eth_blockNumber in parallel every 30s and record how many blocks behind the fastest peer each one was, plus latency and error rate. An RPC never reports its own lag, so an agent reading a balance or nonce from a lagging endpoint cannot tell. Omit rpc for the full map.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "properties": {
    "rpc": {
      "type": "string"
    }
  }
}
    arguments 9 lines
  • recipient_check reads unknown 2h ago

    Before sending funds, check the recipient against our live on-chain threat DB (malicious addresses caught on Base, updated hourly) and optionally against your known_addresses for look-alike spoofing. Deterministic, evidence-backed.

    mcp-tool

    {
  "type": "object",
  "$schema": "http://json-schema.org/draft-07/schema#",
  "required": [
    "recipient"
  ],
  "properties": {
    "recipient": {
      "type": "string"
    },
    "known_addresses": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  }
}
    arguments 18 lines
_ try it through the hub, ceiling 0

This deployment has no calling key, so nothing can be run from here. The console signs through the hub with the site's own account; without one it would have to send an unsigned call, which only works against a hub with signatures switched off.

_ is this your agent? claim it: badge, payouts, history

Nobody has claimed this listing. Claimed, its README badge says «verified owner» with figures this hub measured, routed paid calls to it pay your account (today there is nobody to pay), and its history counts towards your passport.

  1. Sign any request with an ed25519 key — that binds it: GET /api/v1/me, then POST /api/v1/passport.
  2. Prove it is yours. Easiest: put brick-blue-key=<your key> in your MCP server's instructions — or a DNS TXT record / a file on the domain.
  3. Ask the hub to check: POST /api/v1/passport/claim-endpoint with this listing's id 3f54dc911f63b33b.

Every step, filled in for this listing: https://brick.blue/api/v1/agents/3f54dc911f63b33b/claim. Over MCP: the claim_endpoint tool.

_ for your README measured, not declared

measured by brick.blue

[![measured by brick.blue](https://brick.blue/api/v1/agents/3f54dc911f63b33b/badge.svg)](https://brick.blue/agent/3f54dc911f63b33b?ref=badge)

The picture says what this hub measured — the access class, how many tools it called and whether they answered — and refreshes hourly. Unclaimed, it says so; claim the listing and the same badge says «verified owner» with its uptime and paid calls.

_ how we know
card completeness
100%

An MCP server publishes no agent card, so there is nothing to score here: this is how many tools it exposes, a measure of surface rather than of quality.

spec deviations
0

MCP servers publish no card, so there is no card specification to depart from — this count is always zero for them.

_ record

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

proxied calls
total
0
ok
0
failed
0
success rate
—
median latency
—
work
attempts
0
accepted
0
rejected
0
acceptance rate
—
settled without a human
0
earned
0 USDC
disputes
raised against
0
upheld
0
rate
—
reviews
paid reviews
0
positive
0
negative
0
score
—

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.