Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Liquid Agent

REMOTE · API.LIQUIDAGENT.AI

Money tools for AI agents over x402: USDC bridge to 16 chains, Polymarket data, stocks, USDC gas.

0 this week 87 Trust /100

Last scanned (17 hours ago)

Scores higher than 91% of remote servers

Reachable. Last reached 21 Nov 2026.

Category Crypto and Web3

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. How we score → Why this is hard to score →

Endpoint Security80
  • The endpoint's TLS certificate is valid, in date, and uses a strong key. View diagnostics → Pass
  • No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation. See how to fix → View diagnostics → Partial
  • HTTPS is enforced; there's no plaintext access path. View diagnostics → Pass
  • The HSTS (Strict-Transport-Security) header is present. View diagnostics → Pass
  • DNSSEC check failed: this domain isn't protected by DNSSEC. See how to fix → View diagnostics → Fail
Transport & Reachability100
Schema Quality & AI Usability81
  • 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 7558 tokens (~148/item across 51 items; 47 tools + 4 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management100
  • No destabilizing schema changes in the last 30 days.Pass
Tool Coverage100
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 100% of tool parameters carry a description.Pass
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety75
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "stocks_send" implies "send" and declares readOnlyHint instead, contradicting what its own name says it does. See how to fix → Fail
  • An AI judge read all 49 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (13)

  • TLS certificate
  • Authorization
  • HTTPS
  • HSTS header
  • DNSSEC
  • Transport
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Does not apply to this hosted endpoint (2)

  • Supply Chain SecurityScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.
  • Provenance & TransparencyScored on package channels only. This is a hosted endpoint, so there is nothing to install or inspect.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 19 Nov 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Schema quality: 6444 → 7558 ▼ functional
    • Server version: 1.11.0 → 1.13.0 functional
    • New tool “api_audit” functional
    • New tool “api_diagnose” functional
    • New tool “api_test_spec” functional
    • New tool “malware_scan” functional
    • New tool “malware_scan_batch” functional
    • New tool “malware_scan_repo” functional
  • 18 Nov 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Server version: 1.10.0 → 1.11.0 functional
    • New tool “rpc_read” functional
    • New tool “web_extract” functional
    • New tool “web_search” functional
  • 13 Nov 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Tool “link_check” rewrote its description, which is the text the model reads security
  • 11 Nov 26 0
    • Stability: 0.97 → pass security
  • 10 Nov 26 0
    • Stability: pass → 0.97 functional
  • 9 Nov 26 0
    • Stability: 0.97 → pass security
    • The server rewrote its instructions, which are the text every model session reads security
    • Tool “link_check” rewrote its description, which is the text the model reads security
  • 8 Nov 26 0
    • The server rewrote its instructions, which are the text every model session reads security
    • Stability: pass → 0.97 functional
    • Server version: 1.9.0 → 1.10.0 functional
    • New tool “link_check” functional
  • 6 Nov 26 0
    • We updated how we score, so this day's move reflects our rubric, not a change to the server See what changed → functional
Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Probed https://api.liquidagent.ai/mcp

TLS valid

Negotiated TLS 1.3 with TLS_AES_128_GCM_SHA256 .

Subject Issuer Valid from Valid until Key Signature Serial
CN=api.liquidagent.ai CN=YR1,O=Let's Encrypt,C=US 5 Oct 2026 3 Jan 2027 RSA 2048 SHA256-RSA 5a861beaff9940b865129fd138251c3a132
SANs: api.liquidagent.ai
CN=YR1,O=Let's Encrypt,C=US (CA) CN=Root YR,O=ISRG,C=US 3 Sept 2025 2 Sept 2028 RSA 2048 SHA256-RSA a20253f15f2691c05dc1ce13b9bcca4e
CN=Root YR,O=ISRG,C=US (CA) CN=ISRG Root X1,O=Internet Security Research Group,C=US 13 May 2026 2 Sept 2032 RSA 4096 SHA256-RSA f24b6d17f9d9ad7cb1c9fea78782699f

Background: What to check on a remote MCP endpoint →

DNSSEC insecure

Validation of api.liquidagent.ai. — Not signed

Zone DS Keys Algorithms Outcome
. Trust anchor 20326, 38696 8, 8 Verified
ai. DS published 1814 8 Verified
liquidagent.ai. No DS Unsigned (proven) parent-signed NSEC/NSEC3 proves an unsigned delegation
Authentication No authorisation required

The endpoint answered without asking for a token. Anyone who knows the URL can reach it.

Result No authorisation required
HTTP status 200
Header Value
strict-transport-security max-age=31536000; includeSubDomains; preload
content-security-policy default-src 'self'; script-src 'self' 'unsafe-inline' 'unsafe-eval'; style-src 'self' 'unsafe-inline'; img-src 'self' data: https:; font-src 'self' data:; connect-src 'self' wss: https:; frame-ancestors 'none'; base-uri 'self'; form-action 'self'
x-content-type-options nosniff
x-frame-options DENY
referrer-policy strict-origin-when-cross-origin
permissions-policy accelerometer=(), camera=(), geolocation=(), gyroscope=(), magnetometer=(), microphone=(), payment=(), usb=()

Background: How OAuth 2.1 works in the 2026 MCP spec →

Transports 2 probes
Transport URL Outcome Status Location
streamable-http https://api.liquidagent.ai/mcp Verified 200
http (plaintext) http://api.liquidagent.ai/mcp HTTPS enforced 301 https://api.liquidagent.ai/mcp
MCP tools · 47 exposed · ~6,571 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
stocks_set_weights ~137

Free. Sets your vault's target allocation across the 4 stocks in basis points (sum 10000), e.g. [4000, 3000, 2000, 1000]. Default is equal weight. Returns an UNSIGNED transaction {to, data, value, chainId} on Base; nothing happens until you sign it with your own wallet and broadcast it. Liquid never holds keys or funds.

NameTypeReqDescription
vaultstringyesYour vault address.
weightsBpsarrayyes4 weights in basis points summing to 10000, in the basket's constituent order (see stocks_basket).
NameTypeReqDescription
afterstring–What to do after it confirms
attributionobject–ERC-8021 builder attribution suffix info
chainIdnumber–Chain id (8453 = Base)
datastring–Calldata, sign and broadcast as-is
modestring–permit: sign typedData first
nextstring–What to do next
stepsarray–When two transactions are needed (approve, then deposit), sign and send them in order
tostring–Contract to call
typedDataobject–EIP-712 permit to sign off-chain (free, no gas)
valuestring–Native value, always 0

No examples provided.

stocks_signals ~115

PAID over x402 inside this tool call ($0.007 USDC on Base, Polygon or Arc): the whole stock basket's rebalancing signal in one call (returns, volatility, RSI, trend, relative strength and inverse-volatility target weights for NVDA, META, AAPL, GOOGL). Call once without payment to get the x402 PaymentRequired, then call again with the signed payment in _meta["x402/payment"].

NameTypeReqDescription
vaultstring–Optional vault to tailor the signal to.
NameTypeReqDescription
_paidobject–x402 receipt
asOfstring–When the signals were computed
assetsarray–Per stock: returns, volatility, RSI, trend, relative strength
correlationobject–Correlation matrix
notestring–Notes
weightsobject–Inverse-volatility target weights

No examples provided.

stocks_signals_prepare ~85

Returns the x402 payment request for basket signals ($0.007 USDC on Base, Polygon or Arc): rebalancing signal, momentum and suggested weights for NVDA/META/AAPL/GOOGL in one call. Pay it with your own x402 client. This tool does not pay.

NameTypeReqDescription
vaultstring–Optional vault to tailor the signal to.
NameTypeReqDescription
alternativesarray–Other accepted networks
bodyobject–JSON body to send, for POST
howToPayarray–Step-by-step signing instructions
methodstring–HTTP method
notestring–Notes
urlstring–URL to call with the PAYMENT-SIGNATURE header
x402object–The x402 payment request (scheme, network, asset, payTo, amount, EIP-712 domain)

No examples provided.

stocks_vault ~42

Free. A basket vault's value (NAV), holdings, current vs target weights and whether it needs rebalancing.

NameTypeReqDescription
addressstringyesVault address.
NameTypeReqDescription
constituentsarray–Holdings with current weights
mintFeeBpsnumber–Mint fee
navnumber–NAV in dollars
navUsdcstring–Net asset value (USDC, 6 decimals)
ownerstring–Owner
rebalanceBandBpsnumber–Rebalance band
rebalanceNeededboolean–True when weights drifted past the band
strategiststring–Strategist
totalSupplystring–Shares outstanding
vaultstring–Vault address
weightsBpsarray–Target weights in basis points

No examples provided.

usdc_balances ~88

Free. USDC balance (and native gas balance) of an address on every CCTP chain at once: Ethereum, Avalanche, OP Mainnet, Arbitrum, Base, Polygon PoS, Unichain, Linea, Sonic, World Chain, Arc. Use it before paying or bridging to see where your money is.

NameTypeReqDescription
addressstringyesEVM address.
NameTypeReqDescription
addressstring–Address
chainsarray–Per chain: USDC and native gas balance
hintsarray–What to do next
totalUsdcstring–USDC across all chains

No examples provided.

web_extract ~107

PAID over x402 inside this tool call ($0.002 USDC on Base, Polygon or Arc): turn any public web page into clean Markdown. Built to feed the Website Watcher: search with web_search, extract the page, then scan it with link_check. Call once without payment to get the x402 PaymentRequired, then call again with the signed payment in _meta["x402/payment"].

NameTypeReqDescription
urlstringyesPublic http(s) URL to turn into Markdown.
NameTypeReqDescription
_paidobject–x402 receipt
markdownstring–The page as clean Markdown
titlestring–Page title
urlstring–The page URL

No examples provided.

web_search ~105

PAID over x402 inside this tool call ($0.005 USDC on Base, Polygon or Arc): web search with ranked results and LLM-ready excerpts. Pair with web_extract for full-page Markdown, then link_check to scan what you found. Call once without payment to get the x402 PaymentRequired, then call again with the signed payment in _meta["x402/payment"].

NameTypeReqDescription
qstringyesSearch query, e.g. 'pulsechain validator price'.
NameTypeReqDescription
_paidobject–x402 receipt
querystring–The search query
resultsarray–Ranked results: title, url, publish date, excerpts

No examples provided.

x402_check ~193

Free. Decode any x402 payment request (from a URL, a PAYMENT-REQUIRED header or a 402 body) into plain English with warnings, and optionally verify a signed payment (PAYMENT-SIGNATURE / X-PAYMENT header) the way a facilitator would: recipient, amount, time window, EIP-712 signature against the token's real domain, nonce reuse and balance. Every failed check comes with the fix.

NameTypeReqDescription
bodyobject–JSON body for a POST url.
methodstring–HTTP method for url (default GET).
paymentstring–Optional signed payment header value to verify.
requirementstring–Instead of url: the PAYMENT-REQUIRED header value (base64) or the 402 JSON body.
urlstring–The paid endpoint URL (https). It is called once, without payment, to read its 402.
NameTypeReqDescription
decodedobject–The decoded payment request or payment
fixstring–How to fix it
problemsarray–What is wrong
validboolean–Whether the 402 or payment is valid

No examples provided.

Install

How do I install the Liquid Agent MCP server?

Liquid Agent is a hosted endpoint at https://api.liquidagent.ai/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

remote · api.liquidagent.ai

# add to Claude Code
claude mcp add --transport http ai-liquidagent-api-liquid-agent 'https://api.liquidagent.ai/mcp'
Common questions

What is the Liquid Agent MCP server?

Liquid Agent is an MCP server listed in the public MCP registry as ai.liquidagent.api/liquid-agent. Money tools for AI agents over x402: USDC bridge to 16 chains, Polymarket data, stocks, USDC gas. This page covers its hosted endpoint (https://api.liquidagent.ai/mcp).

Is the Liquid Agent MCP server safe to use?

Liquid Agent scores 87 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the Liquid Agent MCP server expose?

Liquid Agent exposes 47 tools: bridge_quote, bridge_prepare_payment, bridge_routes, bridge_status, bridge_referral_earnings, and 42 more. Their descriptions and schemas cost roughly 6,571 tokens of context every time the server is loaded.

Does the Liquid Agent MCP server require authentication?

No. We connected to Liquid Agent without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

Is the Liquid Agent MCP server still maintained?

Liquid Agent is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What did VerifyMCP test on the Liquid Agent MCP server?

VerifyMCP ran 13 checks on Liquid Agent, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.