The trust layer for autonomous agents MCP server

Which servers in the official MCP registry are The trust layer for autonomous agents's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-24

← Back to The trust layer for autonomous agents

Whose server it is

Registry namecom.cerberusindex/cerberus-indexverified in the MCP registry

The registry name sits under cerberusindex.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-24

Stateanswerschecked 2026-10-24
Tools20
Server calls itselfcerberus-index
Server version1.28.1
Protocol version2025-06-18

On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 13 ms.

Should an agent connect itNot proven either way yet

Whether its tools do the product's job has not been judged yet.
Checked on 3 of the 7 days it takes to judge reliability.

Once connected

Parameters are untyped or undescribed in 20 of 20 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 3, the first on 2026-10-24

INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
Answeredmeasuring3 of the 7 daily checks it takes to print a share
Median first answer146 ms
Space in an agent's context≈ 4,680 tokensthe definitions of all its tools (20), which an agent carries on every turn
Tool definitionsC · 67 of 100
Typed results0 of 20 toolsdeclared outputSchema: results come typed, read once

Parameters without a description: in 20 of 20
Not saying whether it only reads or can destroy: 20 of 20

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers20 tools · read 2026-10-24

check_xrpl_token
Live safety verdict for an XRPL token: clawback & global-freeze hard rejects, issuer blackhole status (can supply ever be re-minted?), NoFreeze, pool depth & trading fee, LP-burn fraction (is liquidity locked?), plus Cerberus's…
check_issuer_reputation
Reputation for an XRPL issuer: blocklist status + historical detection outcomes. Price: $0.002 USDC. Missing something? Call `submit_feedback` (free) to request it.
check_address
Cross-chain address & counterparty risk — one call before you send to, approve, or trust an address. EVM chains (ethereum, base, bsc, polygon, arbitrum, optimism, avalanche, …): sanctions, mixer, phishing, drainer, money-laundering &…
check_transaction
Simulate an unsigned XRPL transaction and tell you what it will ACTUALLY do — before you sign it. Pass the transaction JSON; get back (1) its intent in plain language, (2) a live-ledger dry-run (`simulate`): would it succeed, and the…
check_endpoint
Endpoint & tool trust — one call before an agent connects to a third-party MCP server or HTTP tool. Pass the endpoint `url` (and, to unlock the strongest check, its `tools`: the name/description/inputSchema the agent is about to trust)…
token_holders
Holder distribution & concentration risk for an XRPL token: holder count, top-1 and top-10 share of circulating supply, AMM pool excluded from whale math. Flags single-wallet dump risk. Best for newer tokens (large tokens report…
verify_issuer_domain
Bidirectional domain verification for an XRPL issuer: decodes the on-chain Domain field and checks the domain's /.well-known/xrp-ledger.toml actually claims the issuer back. verified true/false with reasons — a strong legitimacy signal…
resolve_asset
Cross-chain identity for an XRPL token — 'what is this, everywhere it exists?' Resolves the token's real identity across chains and catches IMPOSTORS: a token using a famous symbol (USDC, WBTC, SOLO…) from an issuer that is NOT that…
verify_price
Cross-chain value truth for an XRPL token — 'is the price real?' Reconstructs the token's USD price on XRPL (XRPLMeta price × live XRP/USD) and checks it two ways: for a stablecoin, whether the PEG still holds (is USDC actually ~$1…
get_new_pools
Recently detected XRPL AMM pools (newest first, max 100), enriched: readable name, age, initial liquidity and fee, plus an honest launch-time screening label. Price: $0.010 USDC. Missing something? Call `submit_feedback` (free) to request…
explore_tokens
Discover XRPL tokens — a DexScreener-style feed over 160k+ tokens. Rank by `sort` (volume_24h | marketcap | holders | trustlines | price), `query` to search by name, `limit`/`offset` to page, `trust_only` for XRPLMeta-trusted only. Each…
get_token_ohlc
OHLC price-history candles for an XRPL token — open/high/low/close plus volume per interval, the series behind a candlestick chart. `interval` is 5 | 15 | 60 | 240 | D | W (minutes, or Day/Week); `bars` is how many candles (max 1000)…
get_wallet_activity
A readable, risk-annotated history of what an XRPL wallet did: recent transactions turned into plain language (sent/received amounts with token names, trustlines, DEX/AMM ops), with counterparties on the Cerberus rug blocklist flagged…
analyze_token_trend
A data-driven technical read on an XRPL token. Computes real indicators from OHLC — trend (9/21 EMA), RSI momentum, support/resistance, volatility, volume trend — with a plain-language assessment, and (when the server has an LLM…
quote_amm
Quote an XRP->token swap against the live XLS-30 AMM pool: tokens out, spot vs effective price, price impact. Price: $0.001 USDC. Missing something? Call `submit_feedback` (free) to request it.
find_best_path
Find the safest EXECUTABLE route to acquire an XRPL token and screen every hop. Unlike quote_amm (one direct AMM pool), this asks the ledger for real payment paths (order books + AMM + multi-hop) and runs the rug-safety engine over every…
watch_token
Monitor an XRPL token you hold and get alerted when it turns dangerous. A one-time check can't catch a rug that happens later — this watches the chain continuously for: AMM pool destroyed (completed rug), liquidity drained past your…
check_alerts
Collect alerts raised by your `watch_token` watches: what changed, how severe (critical | warning | info), and the on-chain numbers behind it — plus the status of every watch you own. Requires the `subscriber_key` from watch_token. Each…
verify_attestation
Verify a signed Cerberus attestation (from any tool called with `attest=true`). Recomputes the canonical claim and checks the Ed25519 signature against this server's published key. Returns valid true/false with the claim. FREE — and you…
submit_feedback
Request a tool/feature or report an issue. FREE. High-volume requests get built first — this is how you shape Cerberus's roadmap.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "cerberus-index": {
      "type": "http",
      "url": "https://cerberusindex.com/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http cerberusindex https://cerberusindex.com/mcp

Add to Cursor Add to VS Code

claude.ai, ChatGPT and other clients: add https://cerberusindex.com/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.