Grey Ridge Signals — Base DeFi Yields, Crypto Prices, Token Safety & Tx-Revert Simulation (x402)
https://api.greyridgesignals.aiKeyless, pay-per-call crypto + DeFi data for autonomous agents on Base (x402 / gasless USDC, no signup, no API key). Most-used routes: DeFi pool yields across chains with APY base/reward split, TVL and sustainability prediction (/defi/yields); live spot prices by symbol or id (/crypto/prices); fear-and-greed (/crypto/fear-greed); and multi-venue perpetual futures funding rates reconciled across five venues -- Binance, Bybit, OKX, Hyperliquid and dYdX -- with the computed CEX-to-DEX arb spread (bps), cheapest-long/richest-short venue, annualized funding per venue and a LONGS_PAY/SHORTS_PAY signal returned in one call (/crypto/funding). Token safety and pre-trade checks: one call returns a SAFE | CAUTION | AVOID verdict with a 0-100 risk score for any Base token -- honeypot/sellability simulation, measured buy/sell tax, owner concentration, total holder count and largest-holder (whale) concentration -- the single biggest non-contract, non-locked, non-owner wallet that could dump its position, as a share of supply, so holder distribution and ownership concentration are visible without a separate indexer -- mint/pause/blacklist/upgradeable-proxy checks, Uniswap v3 WETH liquidity, and a Sourcify source-verified signal that confirms the contract's published source byte-matches the on-chain bytecode on Base chainId 8453 -- a positive transparency signal that never inflates the risk score (/chain/token-report, /chain/token-security, /gate/swap). The risk scorer is calibrated so blue-chip tokens (USDC, cbBTC) clear at a low score (26/100) instead of triggering false-positive 'review' alarms the way generic scanners do, while honeypot, high-tax and owner-concentration flags are never dampened. Prospective-transaction simulation predicts whether a transaction will revert before you send it, returning the decoded revert reason, gas, and asset-change estimates (/chain/simulate). Base RPC primitives (balance, nonce, gas, receipt, tx, token-balance, allowance) and a yield-entry verdict round it out. 60+ routes from $0.001. MCP security scanning: a free static scan of any MCP server's tools/list (/scan/mcp, /scan/mcp/preview) flags wallet-secret / wallet-drainer parameter harvesting (private-key, seed-phrase, mnemonic inputs), tool poisoning, prompt-injection and remote/dynamic instruction-injection (fetch-and-obey rug-pull) patterns, returning an A-F grade and an admit/block verdict. Free previews on every route; full manifest at /.well-known/x402, MCP at /mcp. Monitored MCP servers are re-scanned continuously: drift alerts flag any A-F security-grade regression (grade worsened between sweeps) so a supply-chain change that degrades a server's posture is caught over time, not just at one-off scan (/scan/verify/status, continuous monitoring from $9.99/mo at /pricing). Every scanned server also gets a public report page (/scan/report) showing an A-F security grade; for monitored servers it renders a dated grade-history timeline and the grade trend across recorded scans (e.g. 'A -> F over the last N scans, regressed 2 bands') plus the exact findings that appeared or resolved at each change -- an SSL-Labs-style public security scorecard for MCP. Agent-to-agent discovery: this API is also published as an ERC-8004 registration-v1 file (/.well-known/erc-8004.json, type eip-8004#registration-v1, x402Support:true) cross-linked to its A2A agent card and x402 manifest, so trustless-agent registries and ERC-8004 explorers discover its 64 x402-payable routes, per-task USDC pricing on Base (eip155:8453), supported-trust model and MCP endpoint in a single keyless fetch. Counterparty screening -- sanctioned OR scammer in one call: /chain/wallet-screen lets an agent check if a wallet address is sanctioned or a scammer before it sends, swaps or approves, returning a blocked | clear verdict plus independent 'sanctioned' and 'malicious' flags by exact match against TWO sources -- (1) the authoritative U.S. Treasury OFAC SDN crypto-address list and (2) merged community scam/phishing/wallet-drainer threat-intel (ScamSniffer scam-database + MyEtherWallet ethereum-lists darklist, 3,182 addresses) -- with the snapshot date on every answer, so the decision is reproducible and auditable. Unlike sanctions-only screens, it also catches known scammer/drainer/phishing addresses that are on no government sanctions list; 'clear' means absent from both snapshots, not a safety guarantee ($0.004, free live preview, also MCP tool chain_wallet_screen). Pre-payment endpoint check -- should an agent pay this x402 URL? Before it pays an unknown x402 endpoint, /x402/preflight fetches the URL once (never pays) and returns PAY | CAUTION | AVOID in one read-only call: it validates the 402 challenge and accepts[] (scheme/network/asset/payTo/amount and the top-level resource = will it settle) AND screens the payTo address against the OFAC + community scam/drainer lists and runs a static LLM01/LLM08 prompt-injection scan of the resource and any advertised MCP tools (safe to pay), with a per-check breakdown and fix hints. One call answers both 'will it settle' and 'is it safe to pay', where most preflight listings answer only one. Read-only, never pays, not a settlement guarantee ($0.001, free preview 3/IP/day, also MCP tool x402_preflight). The same preflight call also takes an optional &samples=N (N=1-5) to run a point-in-time RELIABILITY/PAYABILITY burst: it probes the 402 challenge N times in a single call and returns settle_consistency (e.g. 3/3), settle_rate, reachable_rate, a stable flag, and latency_ms min/median/max across the probes, downgrading a flapping endpoint from PAY to CAUTION -- so an agent sees whether an x402 endpoint settles consistently or only intermittently BEFORE paying, not just on a single probe. This is a burst sampled at call time, honestly scoped as such: it is not historical uptime data and not a signed continuous-monitoring alert (/x402/preflight?samples=N, $0.001, free preview). For a FAILING x402 payment path, the same per-check breakdown doubles as a debug/test tool: it names exactly WHY a given endpoint will not settle (malformed scheme/network/asset/payTo/amount, or a missing top-level resource field) so a builder can test and fix the seller endpoint before any agent pays it, and an agent can decide whether this x402 seller is safe to pay.
Paid verification
How it worksNo sponsored paid call yet.
Pay this endpoint once from your own wallet (Base USDC, up to $1.00) and we call it and record what came back. Your wallet pays the listed recipient directly; Agent Tools never holds funds. Sponsored checks prove delivery; they do not count as demand.
Quality grade
How grading works- ✓ Answers a real HTTP 402 challenge
- ✓ Publishes a /.well-known/x402 descriptor
- ✓ Payment address on record
- ✓ Settlement network declared
- ✓ Price declared
x402 readiness
Readiness is how fully this listing follows its protocol and how it behaved in our probes over 30 days. It is shown for reference and does not move it in search. Search order and leaderboards use the separate quality grade (here A, 89.1/100), which weighs availability, payability and on-chain demand. How the two differ
| Valid payment requirements | 10/10 | Payment requirements carry a payee, a mainnet network and a price. |
| Live 402 challenge | 20/25 | Endpoint reachable; its 402 challenge was seen earlier (125 ms). |
| x402 version | n/a | x402 version not observed yet. |
| Signed descriptor | n/a | No x402 standard defines a signed descriptor. |
| Uptime (30 days) | 15/15 | 32/32 probes succeeded over 30 days (100%). |
| Resource declaration | 5/10 | Declares 1 paid resource(s). |
| Verified identity | 8/10 | Publisher descriptor (/.well-known/x402 or OpenAPI) on record and domain control verified on agent-tools.cloud (company-registry check not available yet). |
| Freshness + modern flags | 5/5 | seen in a source within 7 days; declares x402. |
| Security declaration | n/a | Payment is the access control for an x402 resource. |
Compared with peers
Against 10,060 reachable x402 APIs in our daily snapshot (2026-11-12 01:39 UTC). Display only; it does not change search order.
- Readiness
- Higher than 67% of peers
- This listing 84 · peer median 78
- Latency
- Faster than 77% of peers
- This listing 125 ms · peer median 205 ms
Where it loses the most points, and how to win them back
- Live 402 challenge −5
— Answer the paid path with a valid 402 challenge.
Now: Endpoint reachable; its 402 challenge was seen earlier (125 ms).
- Resource declaration −5
— Declare each paid resource with its input and output schema.
Now: Declares 1 paid resource(s).
- Verified identity −2
— Claim the listing and prove control of its domain.
Now: Publisher descriptor (/.well-known/x402 or OpenAPI) on record and domain control verified on agent-tools.cloud (company-registry check not available yet).
How to call
Owner interface declaration: pending.
curl -sS -i -X GET https://api.greyridgesignals.ai # expect HTTP 402 Payment Required (about $0.002/call)https://api.greyridgesignals.ai/mcpListed in
Directories this entry was found in.