Probing the sell-side book…
ExitGuard is the seatbelt your trading agent calls before it signs. One call proves whether it can actually get OUT at size — and BLOCKs the trades where its own unwind is the market.
pay-per-call · $0.02 USDT0 on X Layer via x402 · MCP-native
The aggregator quote looks clean, so the agent sizes in. Entering is never the problem.
Sell-side depth is thin — often one pool. Nobody told the agent the way out is a keyhole.
Its own unwind moves the market against it. A nominal win becomes a trapped, illiquid bag.
Any autonomous agent hits exit_liquidity_check(token, size) as a pre-trade gate — MCP-native.
Sell quotes laddered at 0.25×–2× your size + on-chain pool reserves. Not a repriced buy quote.
BLOCK / WARN / OK with realizable exit, slippage, % of book, and the raw depth curve on screen.
$0.02 in USDT0 on X Layer via x402. Gas-free. The verdict releases on 200.
Security scanners flag a malicious counterparty. Exit-liquidity is a different failure: an honest market that is simply too thin for you specifically. A DEX aggregator returns a price — not a verdict. Slippage tolerance is a per-fill cap, not a pre-trade gate. The moat is the microstructure model, not the quote.
Est. averted unwind loss on blocked trades: $128k — illustrative, derived from realized-vs-naive slippage on BLOCK verdicts. No vanity counter: value is complete in a single legible call.