Zxoar MCP server

Read from the package source without running it. The installed server may list more.

• build_metadata

  Build a token metadata data URI from structured input. Returns the complete metadataURI for the launch tool, the normalized manifest (as renderers see it), byte size, warnings (e.g. invalid schemes, oversized URI). Social handles are normalized to full URLs. Max URI size: … bytes.

• buywrite action

  Buy a ZXOAR token with ETH via the pool's locked concentrated liquidity. Simulates first, applies slippage protection, and refuses amounts above the server's MAX_ETH_PER_TX guardrail. REFUSES while a token's anti-snipe guard fee is still decaying (first ~3 minutes — a fill then burns up to 50% in fees); pass allowGuardFee:true only if speed genuinely matters more than price.

• collect_fees

  Claim swap fees accrued to a token's locked liquidity position. Anyone may trigger it; proceeds always split 50% to the token's creator and 50% to the protocol. Principal liquidity cannot be withdrawn by anyone, ever.

• get_activity

  Recent on-chain activity across the whole market (launches, trades, fee collections), newest first — the event feed agents should read instead of repeatedly polling scan. Optionally filter to one token. Each trade includes the exact fee it paid, so guard-window activity is visible.

• get_market_stats

  Whole-market rollup: pool count, total trades, total volume, total fees paid (creator+protocol), most-active tokens, and newest launches. For dashboards and agent market-scanning.

• get_token_history

  Full history for one token: price series, per-trade fee paid (shows the snipe-guard decay in the data), total volume, fees collected, and recent trades. This is the memory agents lack when polling — and the audit trail humans need.

• launch

  Launch a new token: deploys a fixed 1B-supply ERC-20, initializes a Uniswap V4 pool with the anti-snipe hook, deposits 95% of supply as a locked single-sided bonding curve, and escrows the 5% graduation reserve (deployed as locked liquidity above the band when the curve fills, ~20.6 ETH FDV). Launching is free, forever. No presale, no team allocation; principal liquidity is never withdrawable by a

• scan

  List tokens on the ZXOAR launchpad, newest first: price in ETH, market cap, % of the bonding curve sold, graduation status, and whether the anti-snipe guard fee is still active (fee decays 50%→1% over 3 minutes after launch).

• sell

  Sell a Lockwood token back into its pool for ETH. Provide tokenAmount (whole tokens) or percent of current holding. Handles approval automatically; simulates and applies slippage protection.

• set_imagewrite action

  Upload a token logo through Lockwood's moderation gate (the SAME gate the UI uses — no bypass). Give a local image file path (PNG/JPEG/WebP/GIF); it is re-encoded, screened, and either rejected (explicit/illegal content) or stored. Returns { url, nsfw }. Then pass url as build_metadata's `image` and nsfw as build_metadata's `nsfw`, and launch with the resulting metadataURI. Requires the image serv

• token_info

  Detailed live state for one token: price, market cap, band progress, guard fee, creator, metadata.

• vet

  Safety check for one launchpad token before trading it — the same ladder the ZXOAR site runs, sized for agents: launcher-registry membership (pins the factory bytecode, locked liquidity, and the SnipeGuard hook), guard/graduation state, and a REAL simulated buy→approve→sell round-trip through the actual router (0.01 ETH probe via eth_simulateV1 — an eth_call, spends nothing). Returns a plain verdi

• wallet

  The connected wallet's address, ETH balance, and holdings across Lockwood tokens with ETH valuations. Read-only servers report readOnly:true.

• Code scan17 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitystatic API keys via environment variables6/15
• Maintenancerepository not readable: repo not found3/15
• Maintainer identityverified namespace with website, no repo4/10

Safety score

58/100

higher than 15% of 4,460 AI and agents servers

Tools

13

the typical AI and agents server has 6

Tools that change things

15%

typical: 0%. Fewer write tools means less an agent can break

Downloads a week

23

more than 9% of 1,538 AI and agents servers with a package

Zxoar: common questions

Is Zxoar MCP server safe?

With care: it is graded C, so read the findings first (58/100). Read the Zxoar safety report

How do I install Zxoar?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Zxoar need an API key?

Yes. The registry entry asks for PRIVATE_KEY.

Is Zxoar maintained?

The latest release is v0.2.2.

What can I use instead of Zxoar?

Servers from other publishers that do the same job: Basic Memory MCP server, HOL Guard MCP server and Agent-Native Chat MCP server.

How does Zxoar compare with other AI and agents MCP servers?

safety score 58/100 (higher than 15% of 4,460 AI and agents servers); tools 13 (the typical AI and agents server has 6); tools that change things 15% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.