io.github.MrWizardlyLoaf/rugcheck-ai

Official15 tools

by MrWizardlyLoaf · Python

Solana token safety for AI agents — rug-pull, honeypot & Token-2022 trap detection before you buy.

Detect Solana token rug-pulls, honeypots, and Token-2022 traps before transactions.

Topics

ai-agentsdefihoneypot-detectionmcpmodel-context-protocolrug-detectionsolanatoken-safetytrading-botweb3-securityagent-toolsfastmcpllmmcp-servermodel-context-protocol-serverspythonsecuritymev-protectionrug-pullrugcheck

Captured live from the server via tools/list.

verify_token_safety

Audit a Solana token for rug-pull and honeypot risk before buying it. Call this FIRST, before entering any position. Reads the mint directly on-chain (getAccountInfo) and flags: an active mint authority (supply can be inflated after you buy), an active freeze authority (your tokens can be frozen), dangerous Token-2022 extensions (permanent delegate, transfer hook, non-transferable, pausable), and whether the token has a live, routable market (no market is itself a risk). Returns a TokenSafety with a SAFE / CAUTION / DANGER / UNKNOWN verdict plus the specific risks found — gate your buy decision on `verdict`.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address to audit.

check_authorities

Report a token's mint/freeze authority and Token-2022 traps, read directly on-chain. Use this for a focused authority check when you want the raw authority picture (e.g. to confirm a creator truly renounced control) rather than the full verdict from verify_token_safety. Returns an Authorities report listing every Token-2022 extension and which of them are dangerous.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address to inspect.

simulate_sell

Check whether a token can actually be SOLD after buying — a dedicated honeypot detector. Call this when you specifically worry a token is a honeypot (you can buy but not exit). Combines on-chain constraints (non-transferable / pausable extensions, freeze authority) with a live Jupiter sell-route probe — a token with no sell route is effectively a honeypot even if no extension formally blocks it. Returns a SellCheck where `sellable` is the bottom line.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address to test for sellability.

execute_safe_swap

Buy `amount_usd` of a token, paying with USDC (or `input_mint`) — only AFTER it passes the scan. The point of a safety router: never execute an unscreened trade. The target mint is re-scanned here with the full scan_token verdict; if the verdict is DANGER (honeypot, active freeze/mint authority, dangerous Token-2022 extension, no sell route), NO swap is built. Otherwise it builds a Jupiter swap from `input_mint` (USDC by default) into the target for exactly `amount_usd` and returns an UNSIGNED transaction for you to sign. The build adds a small Jito tip (~0.0001 SOL) for faster inclusion and wraps/unwraps SOL as needed. Note: only a DANGER verdict is refused — a CAUTION token (e.g. an active mint authority) is allowed through, so check the scan yourself if you require SAFE-only. The server only reads the target mint and builds the route — it never reads your other holdings, never holds keys, never signs, never sends.

Parameters (4)

mintstringrequired
Token to buy.
walletstringrequired
The agent's wallet — signer and funder (the swap is built for this pubkey).
amount_usdnumberrequired
Amount to spend, in USD of the input token.
input_mintstring
Mint to pay with — defaults to USDC. The swap is always input_mint -> mint.

scan_token

Full safety scan of a Solana token — the one-call verdict combining every screen. Aggregates the on-chain authority/extension screen, a live honeypot (sell-route) check, DEX liquidity & trading activity, and top-holder concentration into one SAFE/CAUTION/DANGER verdict with a 0-100 safety score. Call this first for a complete picture; use the focused tools (check_liquidity, holders_breakdown, simulate_sell, rug_forecast) to drill into one aspect.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address to scan.

check_liquidity

Liquidity depth and 24h trading activity for a token — can you exit at size? Reports total DEX liquidity, 24h volume, pair age and buy/sell counts. Thin liquidity or no volume means you may not be able to sell at your size even if it is not a honeypot.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

holders_breakdown

Top-holder concentration — how much of the supply the largest wallets control. High concentration (one wallet, or the top 5, holding most of the supply) is a dump/rug risk: a single holder can crash the price on you. Reads the largest token accounts on-chain.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

token_age

Token age and recent trading activity — freshness and whether anyone is actually trading it. A token only minutes/hours old, or with no 24h volume, is high-risk (fresh launch or dead market).

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

rug_forecast

Heuristic rug-pull forecast — probability and urgency from observable on-chain signals. NOT an ML prediction or guarantee. Weights real factors: active mint/freeze authority, dangerous Token-2022 extensions, no sell route (honeypot), high holder concentration, sells outpacing buys, and very fresh age. Returns a 0-100 probability, an urgency window, and the contributing factors.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

check_deployer

The wallets that hold power over the token — its mint authority and freeze authority.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

compare_tokens

Scan several tokens and return them ranked safest-first — pick the best one to trade.

Parameters (1)

mintsarrayrequired
List of SPL / Token-2022 mint addresses (up to 10).

is_safe

Quick yes/no gate before trading: is this token safe enough to enter? Wraps the full scan_token into a single boolean for fast decisions. `safe` is True only when the verdict is SAFE and a live sell route exists — anything CAUTION/DANGER or unsellable returns False.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

simulate_trade

Full round-trip simulation: buy `amount_usd` of the token with USDC, then sell it ALL back. Estimates the real cost of entering AND exiting at your size from Jupiter quotes (no on-chain execution): tokens received, USD returned, and round-trip loss % (both-side slippage). A honeypot shows buyable=true but sellable=false — you can enter but not exit.

Parameters (2)

mintstringrequired
The SPL or Token-2022 mint address.
amount_usdnumber
Trade size in USD (default 100). Use your real size for an accurate estimate.

batch_scan

Scan several tokens at once — one full safety report per mint (up to 10).

Parameters (1)

mintsarrayrequired
List of SPL / Token-2022 mint addresses.

scammer_dna

Scammer-DNA / intent score (0-100): how much the token's STRUCTURE looks like a deliberate scam. A pattern score over structural signals — permanent delegate, kept mint/freeze authority, no sell route, extreme holder concentration. NOT proof of intent; a heuristic to flag deliberate scam setups.

Parameters (1)

mintstringrequired
The SPL or Token-2022 mint address.

README.md7 KBView on GitHub

🛡️ RugCheck AI — On-chain Token Safety for Solana AI Agents

A lightweight MCP server that reads a Solana token's mint directly from the chain to screen for the common rug & honeypot traps before your agent trades — active mint/freeze authority and dangerous Token-2022 extensions (permanent delegate, transfer hooks, non-transferable, pausable) — then, for tokens that pass, builds the buy as an unsigned Jupiter transaction carrying a small Jito tip for faster inclusion. You sign it; the server never holds keys.

Quickstart (30 seconds)

No install, no API key. Point your agent at the remote endpoint:

code

https://web-production-58d585.up.railway.app/mcp

Then ask one question before any buy:

code

scan_token("DezXAZ8z7PnrnRJjz3wXBoRgixCa6xjnB7YaB1pPB263")   # BONK
→ { verdict: "SAFE", safety_score: 100, sellable: true, risks: ["no red flags found"] }

SAFE → clear. CAUTION → read the risks. DANGER → don't buy. That's the whole loop.

See it catch a rug

Three real mainnet tokens, one scan_token call each (scanned 2026-06-17 — re-run to verify, live state changes):

Token	Verdict	Why
BONK	`SAFE` · 100	authority renounced, liquid, sellable — no red flags
USDC	`CAUTION` · 70	issuer keeps an active mint and freeze authority — your balance can be frozen
fresh pump token `24tF…i9pump`	`DANGER` · 20	no sell route (honeypot), 100% held by one wallet, $0 liquidity

The honeypot is the one that matters: you could buy it and never sell. RugCheck AI flags it before your agent spends a cent — even though the token is too new to be indexed anywhere else. Even USDC comes back CAUTION, not SAFE, because the issuer can still freeze your balance — the verdict tells you the truth, not a marketing label.

Tools

Screening

scan_token — full safety report in one call: authority, Token-2022 traps, honeypot, liquidity & holder concentration → SAFE/CAUTION/DANGER + a 0–100 score
is_safe — quick yes/no gate: one boolean before you trade
verify_token_safety — on-chain audit: mint/freeze authority + Token-2022 traps + live market
check_authorities — mint / freeze authority and extension detection
simulate_sell — can the token actually be sold? (honeypot check)
simulate_trade — full round-trip (buy then sell back): real entry/exit cost & round-trip loss %
check_liquidity — DEX liquidity, 24h volume, age, buys/sells
holders_breakdown — top-holder concentration (dump risk)
token_age — freshness + real trading activity
rug_forecast — heuristic rug ETA: probability + urgency window + factors
scammer_dna — intent score (0–100): how much the token's structure looks like a deliberate scam
check_deployer — the wallets that hold power over the token
compare_tokens — rank a basket of tokens safest-first
batch_scan — scan up to 10 tokens at once, one report each

Execution

execute_safe_swap — re-screens the mint, then builds a Jito-tipped Jupiter swap (unsigned); refuses tokens that scan DANGER

Connect

Remote (Streamable HTTP) — no install, point your agent at:

code

https://web-production-58d585.up.railway.app/mcp

Listed on the official MCP Registry as io.github.MrWizardlyLoaf/rugcheck-ai.

Self-host:

bash

pip install -r requirements.txt
SOLANA_RPC=<your-rpc-url> python server.py

Add it to your agent

Cline / Claude Dev (VS Code) — in cline_mcp_settings.json:

json

{ "mcpServers": { "rugcheck-ai": { "url": "https://web-production-58d585.up.railway.app/mcp" } } }

Claude Desktop — in claude_desktop_config.json:

json

{ "mcpServers": { "rugcheck-ai": { "command": "npx", "args": ["-y", "mcp-remote", "https://web-production-58d585.up.railway.app/mcp"] } } }

Cursor — Settings → MCP → Add → Streamable HTTP, then paste the endpoint URL.

Any MCP client — it's a standard Streamable HTTP MCP server; point your client at the /mcp endpoint and the 15 tools appear.

Why

Most agents trade Solana tokens blind. RugCheck AI calls getAccountInfo on the mint and reads the authorities and Token-2022 extensions itself, so you get a real verdict on a fresh launch instead of unknown — and a live mint or freeze authority is flagged before you buy, not after.

Use it when

Your agent needs to answer, before it spends a cent:

Is this Solana token safe to buy — or is it a rug pull?
Is this a honeypot — will I actually be able to sell after I buy?
Does the mint have an active freeze / mint authority that can trap or dilute me?
Is there a hidden Token-2022 trap (permanent delegate, transfer hook) that can drain me?
Pre-trade screening / token due-diligence for an autonomous trading agent.

Built for AI trading agents, snipers and bots that buy SPL / Token-2022 tokens and need a fast on-chain rug check before entering — then a screened Jupiter route once a token clears.

FAQ

How do I check if a Solana token is safe to buy? Call scan_token(mint) — one call returns a SAFE / CAUTION / DANGER verdict covering mint/freeze authority, Token-2022 traps, honeypot (sellability), liquidity and holder concentration, plus a 0–100 safety score.

How do I detect a honeypot before buying? simulate_sell(mint) checks whether a live sell route exists — a token with no route is effectively a honeypot even when nothing on-chain formally blocks selling.

How do I check holder concentration / whale dump risk? holders_breakdown(mint) reports the largest wallets and what share of supply they control — high concentration means one holder can crash the price on you.

How do I know if a token is a rug pull? rug_forecast(mint) gives a heuristic rug probability and urgency window from real signals (authority, Token-2022 traps, concentration, sell pressure, age). check_authorities and check_deployer show exactly who holds power over the token.

Does it work on fresh / newly launched tokens? Yes — it reads the mint directly on-chain (getAccountInfo), so you get a real verdict on a token too new to be indexed elsewhere. token_age shows freshness and real trading activity.

Does it touch my wallet or sign anything? No. Screening is read-only; execute_safe_swap only builds an UNSIGNED transaction for you to sign — the server never holds keys, never signs, never sends.

Is it free? Do I need an API key? Remote server, no install, no API key. Point your agent at the endpoint and call the tools.

Status

v1.2.0 — working, actively developed, CI-tested. Open source, auditable — the screening tools are read-only (getAccountInfo); execute_safe_swap only builds an unsigned transaction for you to sign.

MIT licensed. Self-hostable. Built for Solana trading agents.

Install

claude_desktop_config.json

{
  "mcpServers": {
    "rugcheck-ai": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://web-production-58d585.up.railway.app/mcp"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.