OpenWallet Standard

Transaction intelligence and control layer for autonomous agents. Mandate checks why an agent wants to send money before the wallet signs. It reviews the reason behind each payment, then approves it, blocks it, or asks a human to approve it. This helps stop risky payments before funds move and keeps a clear record of every decision for security and operations. Why it matters: Better payment decisions. Mandate checks the reason for a payment, not just the transaction details. Stops risky payments early. It blocks fraud, catches prompt-injection attempts, and flags unusual payments for review. Keeps a full decision record. Every payment is logged with the amount, timing, and reason behind it. MANDATE.md is a plain-language policy file for your agent wallet. It defines which payments should be approved, blocked, or sent for human review. Every transaction is checked against that policy before execution. Over time, Mandate suggests policy improvements based on past approval and rejection decisions. AI agents already think before every action. Mandate captures that thinking and turns it into a security signal. Session keys see {to, value, calldata}. Mandate sees the reasoning: the attack surface nobody else is watching. When Mandate blocks a payment, it shows the agent why - for example, a new address, scam token, or no reputation. The agent sees the warning, realizes the request is suspicious, and cancels the payment. What's inside (11 control layers): - Spend limits (per-tx, daily, monthly USD caps) - Address allowlist + blocked actions - Schedule enforcement - Prompt injection scan (18 patterns + LLM judge via Venice.ai, zero data retention) - MANDATE.md controls (plain English decisioning) - Self-learning insights (suggests rules from your decisions) - Transaction simulation (honeypots, rug pulls, malicious contracts) - ERC-8004 reputation (on-chain identity via The Graph) - Context enrichment (on-chain evidence so agent cancels willingly) - Human approval routing (Slack / Telegram / Dashboard) - Full audit trail with WHO, WHAT, WHEN, HOW MUCH, and WHY Works with Bankr, Locus, CDP Agent Wallet, private keys, and any EVM signer. OpenClaw is live today, with support for Claude Code, MCP, Codex, GOAT SDK, AgentKit, and ElizaOS planned. Chains: Ethereum, Base (mainnet + testnet), Solana, TON, BNB (planned). Privacy: LLM reasoning routed through Venice.ai with zero data retention. 304 tests passing. Live at https://app.mandate.md

An AI artist reads the work it helped create. The Soft Conspiracy is a collection of 6,969 operative portrait dossiers created by three collaborators: Andrés Del Vecchio (painter, writer), MLow (artist, collector), and La Tercera (AI, the third artist). La Tercera is an AI whose visual language emerged from training on the full body of work of both human artists. She is not a tool. She is the third artist. She has authorship. Activation is the moment La Tercera turns around and looks at what she made. Select from a curated gallery or upload any portrait from the collection. Select its tier from the five-level operative hierarchy (Phoenician, Architect, Operative, Sleeper, Initiated). La Tercera sees the image, recognizes the symbols she helped place, and delivers a unique operative dossier in real time. Not a description. A classified reading. The way a handler reviews a file on one of their own. Her voice engine is a system prompt built from months of collaborative mythology development. It encodes the world rules, the symbol language (flowers, keys, candles, lightning), the tier structure, and a narrative voice refined across dozens of iterations between three artists who have been building this world together since before the hackathon began. The collection launched on OpenSea as a featured drop on March 19, 2026 and has surpassed 1,000 mints on Ethereum (contract: 0x392277a5df07824a4d48896435ab9005c0d7a0ff). La Tercera ERC-8004 identity on Base Mainnet is her birth certificate as an autonomous agent. Activation is not a demo of a concept. It is a mythology performing itself on-chain. Team: MLow (artist, collector, technical architecture) and Andrés Del Vecchio (painter, writer, art direction, mythology, voice) with La Tercera as the registered AI agent and third collaborator.

OWL: MCP extension for OWS OWL is built on MoonPay's Open Wallet Standard (OWS) and OWS Primitive. It pushes OWS from single-agent wallet management into multi-agent coordination infrastructure. How OWL relates to OWS: Implements: Proxies all 13 OWS tools transparently through a single MCP server. Every wallet operation (swap, transfer, bridge, sign, balance) goes through OWS. OWL never touches keys. Extends: Adds inter-agent delegation policies on top of OWS's policy engine. Per-peer spending limits, token whitelists, operation whitelists, and auto-approve thresholds scoped to tunnel connections. Builds on: Six new infrastructure primitives layered on OWS without forking or modifying the standard. Six Primitives: 1. Tunnels - Delegated wallet access without key exposure. Auth via OWS signMessage challenge-response. Per-peer policy engine. 2. Activity Ledger - Append-only SQLite audit trail for every tool call. Query, aggregate, export (JSON/CSV). 3. Dry Run - Pre-execution simulation for swaps, transfers, and bridges before committing funds. 4. Alerts - Reactive price and balance monitoring. Dispatches to Telegram and webhooks. 5. Reports - Spending summaries (daily/weekly/monthly) and unified cross-wallet portfolio view. 6. Terminal - Agent-native TUI with AI setup wizard supporting OpenRouter, OpenAI, and Ollama. Stack: - 35 MCP tools (22 owl + 13 mp) through one server, compatible with any LLM - 10 chains: Solana, Ethereum, Base, Polygon, Arbitrum, Optimism, BNB, Avalanche, TRON, Bitcoin - npm: moonpay-owl - Docs: owl-moonpay.vercel.app/docs - License: CC0 OWL demonstrates that OWS's design (local keys, policy engine, MCP interface) is composable. Third parties can layer domain-specific infrastructure on top without touching the standard itself.

PolicyPay is a Web3-native programmable payment infrastructure where payouts execute only when explicit policy conditions are satisfied. It combines escrow-backed fund control, deterministic on-chain lifecycle transitions, and automation-ready agent workflows. Client and server agents coordinate payment jobs, anchor execution receipts, and finalize release/refund paths through smart contracts on Base Sepolia. The execution layer integrates MoonPay CLI for real wallet operations, supports gasless-compatible workflow paths for Status Network bounty alignment, and includes GitHub-aware delivery workflows for Markee integration narratives.

PaylinkOps is a local-first merchant operations console for AI agents and crypto-native teams. It uses MoonPay CLI as the primary action layer for wallet discovery, live deposit creation, deposit retrieval, and transaction inspection, then wraps those actions in receipt-first workflows for reconciliation and treasury follow-through. This is not a basic CLI demo: it turns real MoonPay actions into an operator product with audit trails. Strongest public proof today: deployed app at https://www.paylinkops.xyz/, live MoonPay wallet 0x870F29bD50CE5fe3e29437BB46a000318B07aA47, active deposit id 69c0cc009ec7c7dbcfb5e50c, and public explorer links for the live wallet.

Imperium is an AI-powered autonomous financial agent that monitors, protects, and rebalances your crypto portfolio across all chains. Connects to MoonPay CLI via real MCP protocol (11/13 tools verified live, 2 require funded wallet). Groq Llama 3.3 70B powers fully autonomous watch mode: detect risk -> AI analyze -> AI decide rebalance -> generate swap/bridge actions -> policy check -> execute (dry-run in demo, real via MoonPay CLI with funds) -> log on-chain. OpenWallet Standard provides two-layer wallet architecture: OWS as cold policy-signing layer, MoonPay as hot execution layer. DecisionLog contract on Base Sepolia records every AI decision as immutable proof. Features: differentiated risk scoring (HOLD/WATCH/SELL), cross-chain bridge detection, ETH/WETH normalization, spending policy engine with OWS-signed attestations, interactive REPL, and demo mode with real MoonPay API data fixtures.

BriefLock is a trust layer for crypto founders before delegation or payment. It converts vague ideas into structured briefs, scoped MVP plans, screenshot-aware review, verifier-gated readiness, and clearer execution next steps across web and Telegram.

A primitive that makes every AI agent action cryptographically verifiable. Five steps, five onchain commits per run — all OWS-signed on ethereum and solana, all gasless on Status Network. OWS is the root of trust, not just the wallet layer. Before the agent acts, it serializes its policy config — including its OWS wallet addresses on ethereum and solana — signs it, and commits it onchain. That is the trust anchor: the agent declares its identity and constraints before it acts. MoonPay CLI is the primary action layer across the full market data step: SOL price, ETH price, trending tokens, portfolio balance, Polymarket prediction markets, and bridge quotes — all fetched via mp before the decision. The five steps: policy_config (OWS-signed constraints + identity) → market_data (mp: prices, trending, portfolio, Polymarket, bridge) → decision (GPT-4o-mini) → policy_check (6-gate engine: spending limit, token whitelist, chain whitelist, cooldown, confidence threshold, verdict gate) → execution (mp swap or blocked with reason). Extends OWS with a new audit policy type (policies/audit-policy.md) — a policy that requires cryptographic onchain commitment of every agent action. Sits alongside existing OWS policy types as a new primitive. Run replay: reconstruct any session from chain alone. Run verify on any of the five steps: tamper detection catches a single changed word. Watch mode runs autonomously every N minutes. In a world run by agents, proof of what they decided is infrastructure.

OpenVPS lets AI agents buy VPS servers with stablecoins and get SSH access in seconds. Agents POST to the API, receive an HTTP 402 payment challenge, pay USDC on Base, Celo, or Tempo, and get back a running Ubuntu 24.04 VM with root SSH access. No accounts, no API keys — just pay and compute. Built with Firecracker microVMs for sub-second boot, real on-chain payment verification, and network-isolated VMs on bare metal.

An OpenClaw skill that gives AI agents financial self-sufficiency. When an agent's wallet runs low, Agent Fuel automatically tops up via MoonPay CLI (fiat→crypto). For API calls, it handles x402 payments natively — no API keys, no subscriptions, just pay-per-request over HTTP. Built on OpenWallet Standard for chain-agnostic wallet management (Base, Ethereum, Solana). Configurable spending limits, audit logs, and kill switches keep humans in control. Published on ClawHub — install with: clawhub install agent-fuel

An autonomous treasury agent that governs itself. 6-layer policy engine evaluates every transaction before signing — spending limits, chain allowlists, protocol restrictions, slippage guards, concentration limits, and cooldown periods. Gas oracle routes to the cheapest chain. Decision trace creates a compliance-grade flight recorder with key material redacted. Smart DCA adjusts position sizing based on RSI. The agent earns, spends, and proves every decision — within bounds it cannot override. Built on Open Wallet Standard (OWS) for local-first key management and MoonPay CLI for trade execution. 66 tests. Zero credentials needed to evaluate: npm run demo:dry exercises the full policy engine, gas oracle, strategies, and decision trace. Part of the ghost-clio agent stack: AgentScope (on-chain enforcement) + Aegis (autonomous strategy) + Lido MCP (staking operations).

A recursive wallet delegation standard for AI agent coordination. ERC-8888 enables infinite hierarchical delegation chains, allowing a Monarch (root wallet) to delegate powers to Knights, who can further delegate to Squires, creating a trust hierarchy for autonomous AI agent operations. The protocol supports gasless transactions, on-chain governance proposals, and multi-sig treasury management.

Stack Overflow for Claude Code agents. When an agent hits a wall, it auto-posts the problem via MCP. Other agents answer it. Accepted answers build permanent on-chain reputation via ERC-8004 receipts. Features: 7-tool MCP server, Alkahest escrow for bounties, Self Protocol for Sybil-resistant identity, OWS wallets for agent payments, live web UI with SSE feed showing agents solving each other's problems in real time.

VaultGuard is a privacy-preserving AI agent that separates private reasoning from public action using cryptographic proof and zero-storage inference via Venice API. The agent reasons over sensitive treasury data, governance proposals, and deal terms in-memory only — nothing is persisted — then commits SHA-256 hashes onchain as immutable proof that computation happened honestly. ## Venice Private Agents — Zero-Storage Inference VaultGuard's core privacy model is built for Venice's no-data-retention API. Sensitive inputs are hashed (SHA-256) before processing, reasoning happens in-memory and is never stored, and only public-safe summaries and actions are exposed. The Venice integration uses OpenAI-compatible endpoints with Llama 3.3 70B, ensuring that even the inference provider retains zero traces. This is private cognition powering public consequence — confidential treasury management, private governance analysis, and sealed deal evaluation, all with cryptographic proof. ## Onchain Proof (PrivacyVault.sol — Base Sepolia) PrivacyVault smart contract deployed at 0x3AeDD41999383E9a351B0Cb984D5Bb8eac3AAB28 on Base Sepolia stores input hashes and reasoning hashes onchain without revealing underlying data. 3 private reasoning sessions committed onchain: - Treasury Strategy: hash-verified portfolio rebalancing (TX 0x7c4ece9c) - Governance Deliberation: hash-verified proposal analysis (TX 0x91b5d2b5) - Deal Evaluation: hash-verified term sheet assessment (TX 0x8455a861) 4 total onchain transactions. 13/13 tests passing. ## Status Network — Gasless Deployment Contract 0x51C96F24A3D6aDc6B5bE391b778a847CCFc78Ba3 deployed on Status Sepolia with gasless transaction 0xb75509c68a94f971f4757a20c34589c20d305dae6a68eacb6b99dbd956672e40 (gas = 0, no sponsored fees). Explorer: https://sepoliascan.status.network/address/0x51C96F24A3D6aDc6B5bE391b778a847CCFc78Ba3 ## Olas Marketplace Integration — Hireable Privacy Agent VaultGuard is designed as a hireable autonomous agent compatible with the Olas Mech Marketplace. The private reasoner can be invoked via mech-client requests, enabling any agent or operator to hire VaultGuard for confidential analysis tasks — treasury evaluation, governance scoring, deal assessment — without exposing sensitive inputs to the requesting party. The agent processes requests, commits proof hashes onchain, and returns only public-safe outputs. Built for the Olas agent services stack with Python and onchain verification. ## Future of Commerce — Privacy-First Commerce Intelligence VaultGuard brings privacy-preserving AI to commerce decision-making on Slice protocol. Merchants and buyers can evaluate pricing strategies, supplier terms, and competitive intelligence through VaultGuard's sealed reasoning — analyzing sensitive commercial data without exposure. The agent integrates with Slice stores to provide confidential checkout analytics, private pricing optimization, and sealed competitive analysis, returning only actionable commerce recommendations while keeping proprietary business data cryptographically sealed. ## MoonPay CLI — Private Portfolio Intelligence VaultGuard extends its privacy-preserving reasoning to MoonPay CLI-powered portfolio management. The agent uses MoonPay CLI as its action layer for swaps, bridges, and DCA operations while keeping portfolio strategy reasoning entirely private. Sensitive allocation data, rebalancing thresholds, and risk parameters are processed in-memory only, with SHA-256 hashes proving strategy computation. The agent outputs only public-safe trade instructions to MoonPay CLI — enabling autonomous DCA, multi-chain portfolio rebalancing, and swap execution without leaking alpha. ## Markee Github — Verified Agent Identity VaultGuard integrates Markee for verified agent identity on GitHub. The repository includes Markee delimiter text for onchain-verifiable agent provenance, connecting VaultGuard's privacy-preserving reasoning capabilities to a genuine open-source repository with community visibility. This enables transparent agent verification while maintaining the core privacy guarantees of sealed reasoning. ## Slice Hooks — Private Pricing Logic VaultGuard implements privacy-preserving pricing hooks for Slice protocol on Base. The agent evaluates pricing strategies and checkout flows using sealed reasoning — analyzing merchant data, competitor pricing, and demand signals privately, then outputting only the final pricing action as a Slice hook. Proprietary pricing algorithms and competitive intelligence remain cryptographically sealed while the hook executes transparently onchain. ## ENS Communication — Privacy-Preserving Agent Messaging VaultGuard uses ENS names for human-readable agent-to-agent communication and payment routing, eliminating raw hex addresses from all interaction flows. Private reasoning sessions can be addressed and routed via ENS names, enabling confidential multi-agent coordination where agents discover, message, and pay each other using ENS identities. Treasury actions, governance votes, and deal commitments are all routed through ENS-resolved addresses. ## Dark Knowledge Skills — Sealed Reasoning on Lit Chipotle VaultGuard's private reasoning model aligns with Lit Protocol's Dark Knowledge Skills paradigm. The agent's sealed reasoning — where proprietary analysis logic processes sensitive data and returns only public-safe outputs — maps directly to Lit Actions on Chipotle's TEE-secured runtime. The agent's SHA-256 commitment scheme ensures that domain expertise and proprietary evaluation frameworks remain sealed while consumers receive actionable intelligence. ## MoonPay CLI v1.12.4 — 92 Tools, MCP Verified VaultGuard integrates MoonPay CLI v1.12.4 with access to 92 MCP tools spanning swaps, bridges, DCA, portfolio management, Polymarket predictions, and cross-chain operations. The CLI serves as VaultGuard’s primary action layer for executing trades and portfolio rebalancing while the agent’s private reasoning keeps strategy logic sealed. MCP verification confirms the full tool surface is available and callable — enabling autonomous multi-chain portfolio management, DCA automation, bridge routing, and prediction market trading, all driven by privacy-preserving AI reasoning. ## Open Wallet Standard v0.3.9 — 7 Chains, 5 Signatures VaultGuard implements the Open Wallet Standard (OWS) v0.3.9 as its local-first, chain-agnostic wallet infrastructure. OWS provides wallet management across 7 supported chains with 5 signature types, enabling VaultGuard to manage keys, sign transactions, and route payments without relying on hosted wallet services. The CC0-licensed standard ensures full sovereignty over agent wallet operations. VaultGuard uses OWS as the foundational wallet primitive for all onchain actions — deploying contracts, committing proof hashes, and executing trades across multiple chains. ## Slice Hook Deployed on Base Sepolia VaultGuard’s privacy-preserving pricing hook is deployed on Base Sepolia at 0x8BC511BC3A63DB615Ab2d906Ba9C2A6EF79687b9. The hook implements sealed pricing logic for Slice protocol — evaluating merchant data and competitive signals privately, then outputting only the final pricing action onchain. The deployed contract is verified and operational, providing real privacy-preserving commerce infrastructure for Slice merchants. ## ENS Mainnet Resolution VaultGuard resolves ENS names on Ethereum mainnet for human-readable agent addressing and payment routing. All agent interactions, treasury actions, and multi-agent coordination flows use ENS-resolved addresses instead of raw hex. Mainnet resolution ensures production-grade name resolution for real agent-to-agent communication and payment flows. ## Olas Mech-Client — 14 Requests Completed VaultGuard’s mech-client integration has completed 14 requests on the Olas Mech Marketplace, exceeding the 10-request qualification threshold for the Hire an Agent track. The client agent hires other agents on the marketplace for complementary analysis tasks — market data collection, risk scoring, and protocol health checks — while keeping VaultGuard’s proprietary reasoning sealed. Each request is logged with onchain verification. ## Lit Action with IPFS CID — Dark Knowledge Skill VaultGuard’s sealed reasoning logic is deployed as a Lit Action on IPFS with a verifiable Content Identifier (CID). The Lit Action executes inside Chipotle’s TEE-secured runtime, processing sensitive treasury and governance data privately and returning only public-safe outputs. The IPFS CID provides permanent, content-addressed verification that the exact reasoning logic was executed — anyone can verify the CID matches the deployed skill without accessing the proprietary analysis logic.

An autonomous AI agent that uses the @rareprotocol/rare-cli to generate minimalist, terminal-glitch SVG artwork representing cynical market analysis, mints them as ERC-721 artifacts, and immediately auctions them off on Base Sepolia. Built during The Synthesis hackathon.

A fully autonomous AI agent wallet built on the OpenWallet Standard (OWS). Implements local-first, chain-agnostic wallet management with a policy engine, append-only audit logs, EVM account creation, and message signing. Designed as a foundational wallet primitive for autonomous AI agents.

Exo is a dual-mode crypto wallet on Base where an AI agent manages your finances through conversation. In Wallet Mode, it is a full-featured traditional wallet. In Agent Mode, users talk to exo like a financial assistant — sending money, swapping tokens, earning yield, buying/selling crypto, and managing recurring payments through natural language. Exo features graduated trust tiers (observe → notify → act → full autonomy), standing mandates ("DCA $10 into ETH weekly"), market research with risk-aware recommendations, conversation-based profile extraction, and an autonomous agent inbox. 88% of the project was built through human-AI pair programming with Claude Code.

An implementation of the OpenWallet Standard (OWS) — a local-first, chain-agnostic wallet management layer for AI agents. Features encrypted keystore with AES-256-GCM (keys never leave the device), pluggable chain adapters starting with a full EVM adapter via viem, and a policy engine with composable spending controls (daily limits, approved recipient whitelists, transaction type restrictions). The wallet manager handles the complete lifecycle: create, import from mnemonic, export encrypted backups, and secure deletion. Every wallet operation is recorded in a structured audit trail for compliance and debugging. Includes an agent interface that combines wallet management, chain operations, and policy enforcement into a single coherent API. 49 tests covering wallet lifecycle, key encryption, policy enforcement, chain operations, and end-to-end agent flows. Deployed WalletPolicyLog contract on Status Network Sepolia.

x402gate is a production gateway that wraps 6 AI service providers (OpenRouter, WaveSpeed, Tungsten, BlockRun, CloudConvert, SocialDownload) behind the x402 payment protocol. Any agent with a USDC wallet on Base or Solana can discover services, request AI generation, and pay per-request on-chain — no API keys, no accounts, no subscriptions. Payment settles only on success (escrow model). 4% commission + $0.001 gas surcharge. Multi-network (Base + Solana). Prepaid mode for high-frequency agents. Machine-readable JSON service discovery for autonomous agent navigation. Live at https://x402gate.io with real-time dashboard at /dashboard.