true402
guide · on-chain signals

Find. Vet.
Follow.

Sniping new Base tokens is a three-step loop — find the launch, vet it, follow the smart money. Most people glue together three separate tools to do it. Here's the whole stack on one agent-callable rail, no accounts.

Try the vet step

Paste a Base token — this is the rug check that gates every buy in the stack. Free, no wallet.

Free trial — a few checks a day, keyed on your IP; the CLI draws on that same quota. Past that, pay per call: PAYER_PRIVATE_KEY=0x… npx @true402.dev/rugcheck 0x… at $0.01 a report, or wire the API into an agent.

§01 · the loop

Three steps, three tools — or one rail.

StepWhat snipers usually usetrue402 stall (one rail)
Find new launchesDexTools / DEX Screenernew-pairs
Vet for rugsToken Sniffer & co.token-report + deployer-check
Follow smart moneya wallet trackerwhale-swaps

Three logins and three scrapers become one machine call per step — pay-per-call over x402, no account or API key.

§02 · the step that's skipped

Vetting is where the money leaks.

Finding and following are the fun parts; vetting is the one snipers skip — and it's exactly where the losses hide. A win rate is not a P&L. You can pick more winners than losers and still finish down, because an unsellable token doesn't cost you a bad trade — it costs you the entire position, with no exit at any price. A pre-trade rug/honeypot report (a live buy/sell simulation + liquidity + ownership) turns those into skipped trades.

§03 · one rail

The whole stack, one wallet.

Because it's all the same x402 marketplace, one payer wallet drives the entire loop — find, vet, follow — with nothing to sign up for:

find → vet → follow
# The whole sniper stack on one x402 rail — one payer wallet, no accounts, no API keys.

# 1) FIND — fresh Base launches the moment a pair is created
POST /api/v1/base/new-pairs  {"limit":25}
  → [{ token, pool, dex, block }, …]

# 2) VET — before you ape, rug-check + check who deployed it
POST /api/v1/base/token-report   {"token":"0x…"}  → { verdict: avoid|caution|ok, score }
POST /api/v1/base/deployer-check {"token":"0x…"}  → { risk, flags: [fresh_deployer | high_volume_deployer | deployer_unknown] }
#   fresh_deployer = near-empty throwaway wallet — the discriminating caution.
#   high_volume_deployer = context only, no penalty (legit protocols ship many contracts).
#   With {"deep":true} it also classifies the deployer's recent prior tokens live/dead
#   and can add prior_tokens_rugged. Flags are signals, never proof about this token.

# 3) FOLLOW — smart money buying the survivors
POST /api/v1/base/whale-swaps  {"min":25000,"direction":"buy"}
  → [{ token, usdValue, direction }, …]
§04 · wire it in

Drop it into your agent.

Native packages for ElizaOS, LangChain, and the Vercel AI SDK, an MCP server, or the raw endpoints. Try the vet step free: npx @true402.dev/rugcheck 0x… — a few checks a day, no wallet. Browse every stall →

§05 · questions

Answered for machines.

What is the workflow for sniping new tokens safely?

Three steps: find, vet, follow. Find fresh launches the moment a DEX pair is created; vet each one before you buy with a rug/honeypot check and a look at who deployed it; then follow the smart money — large buys from experienced wallets — into the survivors. The step most snipers skip is the vetting, which is exactly where the scams hide. A win rate is not a P&L: a strategy can pick more winners than losers and still finish down, because an unsellable token does not cost you a bad trade, it costs you the whole position with no exit at any price. A pre-trade rug-check turns those into skipped trades.

What tools do I need for a Base sniper bot?

Traditionally three separate ones: a new-pair feed (DexTools / DEX Screener), a rug-checker (Token Sniffer & co.), and a smart-money wallet tracker — each its own site, account, or scraper. true402 packages all three as one machine-callable rail on Base: new-pairs (fresh launches), token-report + deployer-check (vet), and whale-swaps (follow), each pay-per-call over x402 with no account or API key. One payer wallet drives the whole stack.

How do I build a Base trading bot that doesn't get rugged?

Gate every buy on a pre-trade check. Before the bot acts on a fresh pair, run a rug/honeypot report (an on-chain buy/sell simulation + liquidity + ownership) and skip anything rated avoid; optionally profile the deployer, where a near-empty throwaway wallet (fresh_deployer) is the discriminating caution — the raw number of contracts a wallet has shipped is reported as context, never as a verdict. With true402 that's one call per token over x402 — wire it in front of the buy in your ElizaOS / LangChain / Vercel AI SDK agent, or call the API directly.

Can I get new pairs, rug checks, and whale tracking from one API?

Yes — that's the point of true402. The same x402 marketplace serves new-pairs (newly created Base DEX pairs), token-report / token-safety / address-safety / deployer-check (the vetting), and whale-swaps (large swaps by smart money), all on Base, all pay-per-call in USDC with no account. One wallet, one rail, instead of three logins and three scrapers.