Skip to main content
Glama
PocketSlip MCP Server

PocketSlip MCP Server

by NARAProtocol

CryptoSLIP TypeScript SDK & PocketSlip MCP (v1.0.0)

CI License: MIT Network: Base 8453

Official Zero-DOM TypeScript 6 SDK (@cryptoslip/sdk) and Bounded Agent MCP Server (@cryptoslip/mcp) for CryptoSLIP Protocol V3.2 on Base Mainnet (8453) — live at https://slip.naraprotocol.io, ENS https://cryptoslip.eth.limo, and Telegram Mini App @cryptoslip_bot/SLIP.


Packages

Package

Path

Description

@cryptoslip/sdk

packages/sdk

12 tree-shakeable zero-DOM subpaths (Node 22+, Cloudflare Workers, Deno, Bun, Browsers): 1-tx Multicall3 batch payouts (1..500 items in <50ms offline signing), zero-key withCryptoSlip() HTTP 402 Hono/Fetch/Express middleware + verifyPartnerWebhookSignature(), 0% 1st Keep in Slips (1:1 non-commingled wallet privacy), Circle TokenPaymasterV08 USDC network fee UserOps, VOPRF encrypted voice notes, verifiable EIP-712 receipts, and BoundedAgentPolicy spend firewall.

@cryptoslip/mcp

packages/mcp

PocketSlip Model Context Protocol (stdio) server & OpenClaw SKILL.md. Provisions AI agents with a bounded # slip URL (<1s boot Keep rotation invalidating the .env link at 0% fee) instead of a raw wallet private key, exposing 7 policy-guarded tools.


Related MCP server: Agent Clearing Desk

Quick Install

From npm (or Direct Signed GitHub Release Tarball)

# Standard install
npm install @cryptoslip/sdk viem

# Or pin directly to the verified v1.0.0 GitHub Release tarball
npm install https://github.com/NARAProtocol/cryptoslip-sdk/releases/download/v1.0.0/cryptoslip-sdk-1.0.0.tgz viem

Why Builders & AI Agents Integrate CryptoSLIP

  1. 1-Tx Multicall3 Partner Dispenser (1..500 Links, Zero ERC20.approve): Mint up to 500 revocable DM links or irrevocable No take-back (cancelLockedUntil = 9_999_999_999) bounties in a single Base transaction via EIP-3009 ReceiveWithAuthorization + MintPermit, while preserving slip.creator == partnerWallet (0% creator reclaim on unopened DM links + 0.25% lifetime on-chain referrerOf revenue share).

  2. 0% 1st Keep in Slips (1:1 Non-Commingled Wallet Privacy): Recipients and agents rotate claimSigner on-chain at 0% protocol fee (nonces(slipId): 0 -> 1), keeping funds in isolated 1:1 positions rather than exposing their primary wallet balance or trade history.

  3. Zero-Key HTTP 402 Tollbooth (withCryptoSlip) & Signed Webhooks: Gate Hono, Fetch, or Express APIs directly into a watch-only multisig Safe with synchronous SlipReplayGuard, stateless X-Payment-Refund-Slip overcharge refunds, and HMAC-SHA256 verifyPartnerWebhookSignature() (slip.claimed, slip.kept, checkout.paid).

  4. Bounded PocketSlip Agent Wallets (@cryptoslip/mcp): Never hand an LLM or untrusted sub-agent your main wallet private key. Hand it a $5.00 USDC slip URL; on startup, PocketSlipVault rotates the key on-chain in <1s and enforces out-of-context BoundedAgentPolicy per-call/daily USD caps, sliding-window rate limits, and output secret scrubbing.


Canonical Base Mainnet (8453) Verified Contracts

Contract

Address

Notes

CryptoSlipVaultV32 (v33)

0x3f80aa68Fb4834d6E119EE1cf932C14419Cf63EB

Active micro-minimum vault (0.10 USDC, 100 sats cbBTC, 1e-6 ETH / NARA)

CryptoSlipNFTV32 (v33)

0x356a701DABDeabe61F7Fc79D730617CEcbd7B94f

Synchronized 1-of-1 position NFT

CryptoSlipTradeRouterV32 (v33)

0x2069CA17D45d3421D28f15F20262575333CECF51

Slip-to-slip & swap-and-pay router

NARALiquidityHookV5

0x07d8F2Dfd121B10dE3176DeAb6A8C0eaa2646aCC

Canonical Uniswap V4 Hook for 100% of NARA swap legs

Base Builder Code (ERC-8021)

bc_ea68lz8s

Appended automatically as Schema 0 calldata suffix


Local Verification

cd packages/sdk && npm install && npm run typecheck && npm run lint && npm test && npm run build
cd ../mcp && npm install && npm run typecheck && npm run lint && npm test && npm run build

Security & Disclosure

License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents to call backend tools under per-agent policy, deciding which tools each agent can see, what arguments it may pass, how often it may call, and which writes wait for human approval. It also redacts secrets and PII from tool results, fences untrusted output against injection, and records every call in a hash-chained audit log.
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables autonomous AI agents to invoke metered upstream tool APIs through a policy firewall that enforces per-key cent-level spending caps and strict tool allowlists, blocking unauthorized or over-budget calls pre-flight. Every transaction is settled atomically and returned with an HMAC-SHA256 signed audit receipt, with instant key rotation and spread accounting built in.
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents to call tools through a security gateway that intercepts requests, enforces deterministic policy and risk checks, blocks or escalates high-risk actions for human approval, and keeps tool credentials and audit logs protected.
      1
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Exposes enterprise tools to third-party AI agents through a single guard that authorizes each call against default-deny, purpose-scoped entitlements, executes it, and writes a hash-chained, tamper-evident decision trace the firm owns. It also verifies signed vendor webhooks and enforces a strict versioned contract, so agent activity stays scoped, auditable, and portable across vendors.
      MIT