Coinbase

Top Builders

Explore the top contributors showcasing the highest number of app submissions within our community.

Coinbase

Coinbase is a leading cryptocurrency exchange and blockchain infrastructure company founded in 2012. The company provides a comprehensive platform for buying, selling, and managing cryptocurrencies, along with developer tools and infrastructure services that enable builders to create innovative Web3 applications. Coinbase's developer platform, Coinbase Developer Platform (CDP), offers APIs, SDKs, and services that simplify blockchain integration and payment processing.

General
Company	Coinbase
Founded	2012
Repository	https://github.com/coinbase

Start building with Coinbase's products

Coinbase offers a wide range of developer tools and services that empower developers to build next-generation Web3 applications. From payment protocols like x402 that enable AI-to-AI payments, to wallet infrastructure and blockchain APIs, Coinbase's developer platform provides the tools needed to integrate cryptocurrency and blockchain functionality into applications. Explore the possibilities with Coinbase's solutions and see what you can create during lablab.ai hackathons.

x402

x402 is an HTTP-native payment protocol that enables AI agents and applications to make programmatic payments without accounts, credit cards, or manual approvals. Built on EIP-3009, x402 turns the unused HTTP 402 status code into a working payment system that allows autonomous agents to discover, understand pricing, and pay for API access automatically. Learn more about x402.

Edit on GitHub

Coinbase AI Technologies Hackathon projects

Discover innovative solutions crafted with Coinbase AI Technologies, developed by our community members during our engaging hackathons.

AgentPayOps Finance controls for autonomous agents

Autonomous agents will buy data, tools, compute, and services. Companies need a control plane that decides when those agents are allowed to spend money, blocks risky or duplicate purchases, escalates high-value actions, and records every decision for finance teams. AgentPayOps demonstrates that control layer through one vertical workflow: An invoice agent reviews a vendor invoice. The agent needs a paid vendor-risk report. The report endpoint returns 402 Payment Required. AgentPayOps evaluates vendor, category, amount, approval threshold, and duplicate-purchase rules. The system approves, blocks, or escalates the payment. Transactions and reasoning are shown in an audit dashboard.

DisburseGuard

DisburseGuard is a proof-paid treasury firewall for autonomous payout workflows. Instead of letting an AI agent summarize risk and move on, DisburseGuard forces the agent to acquire evidence before capital can move. A payout intent enters the system, Gemini extracts invoice and vendor context into structured fields, and a Proof Agent creates a multi-proof marketplace plan across vendor-risk, recipient-match, sanctions-screen, and delivery-attestation resources. Each proof endpoint is protected by an x402-style HTTP 402 payment gate. The Payment/Quote Agent buys deterministic fallback receipts, or stops proof spend early when a hard-risk signal already blocks the payout. Policy Guard then clears, limits, reviews, or blocks the payout based on the receipt set. Audit Agent signs a ClearancePacket containing the decision, approved amount, policy version, proof hashes, expiry, and rationale. Every step is written into a Vultr-backed append-only ledger with event hashes and previous-event hashes, so /api/clearance/:id/verify can prove both packet authenticity and chain integrity. The demo reports proof spend, skipped proof spend, capital requested, capital approved, and capital controlled.

x402 Bazaar Agent

Today's AI agents hit a wall the moment they need data behind a paywall. The x402 Bazaar Agent tears that wall down. The agent takes a plain-English question, searches a catalog of 42 premium endpoints across 17 domains using Gemini semantic embeddings, selects the cheapest APIs that can answer, pays for each call via x402 micropayments on Base Sepolia, and composes a cited final answer — no human in the loop. DISCOVER — Gemini text-embedding-005 indexes the catalog at startup. Queries rank by cosine similarity; entries below a relevance floor return zero results, giving the agent a clean stop signal rather than hallucinated answers. PAY — The x402 HTTP client intercepts the 402 response, signs a USDC micropayment with an EVM wallet, and retries. The LLM never sees the payment handshake. VERIFY — After every paid call, an independent Gemini Flash model judges whether the response served the stated purpose. Off-topic data is flagged before the reasoning model builds on it. COMPOSE — Gemini 2.5 Pro weaves results across endpoints and domains into a cited answer. Demo: "Will container ALPHA-99 arrive at Genoa on time?" triggers three paid calls — container ETA, port congestion, marine weather — producing a synthesized delay forecast. Guardrails prevent runaway behavior: hard budget cap, empty-search streak, no-progress streak, duplicate-call detection, and an iteration backstop. The catalog mirrors the real Coinbase Bazaar's discovery schema exactly — switching to the live Bazaar is one function change. All 42 endpoints have mock fallbacks so the demo works with zero mandatory signups. The browser UI streams every agent action live.

Mai - Agentic commerce for Fashion

Have you bought an outfit that looked great in-store but fell flat at the event? Have you scrolled through page after page of product grids, overwhelmed by options and filters that don't actually understand what you're looking for? Online shopping hasn't fundamentally changed in twenty years. Since the early 2000s, the experience has been the same one-sided transaction: browse a grid, filter by size, add to cart, checkout. You still can't bring your closet. Mai is that shift — from e-commerce to a-commerce. You bring your wardrobe, body, and style into the conversation. This makes online more powerful than in-store — you can't carry your closet to a physical shop, but with Mai you can. Visualize how new pieces work with what you already own. How it works: Upload a portrait. Your AI stylist builds your Style DNA: silhouettes, palettes, style eras & fabrics. Every product is scored across five categories against your profile. Only what genuinely fits you is shown, with transparent reasoning. Try before you buy: Generate try-on images and outfit videos via Veo — in any environment you choose. Pin a location (a restaurant, office, or wedding venue) on Maps and Mai pulls real imagery of that place, then shows you wearing the outfit there. Outfits, not items: Your cart is analyzed as a complete outfit across six categories. The agent proactively detects gaps ("jacket and pants but no shoes") and finds what completes it. Built on a multi-agent architecture with open protocols (A2A, UCP): a client agent represents you, a merchant agent the seller. Payments (via AP2): human-present checkout with cryptographic challenge-response, and human-not-present mandates that let the agent autonomously purchase Wishlist items within spending limits you set. Mai is hyper-personalization for fashion. Shifting fashion retail from CRM to CAM — Customer Aura Management: understanding how you feel about what you wear, what makes you confident, and what would actually work for your life.

TrapLedger

Autonomous AI agents are increasingly capable of initiating real payments on behalf of users and enterprises. TrapLedger exists to answer a critical question no one is asking yet: what governs an agent's payment before the wallet signs? TrapLedger sits between an Agent Client and a Paid Resource. When an agent initiates a Payment Attempt, TrapLedger first fetches a Payment Challenge (x402-shaped 402 response) from the target resource, then applies a deterministic Policy Set, destination allowlist, maximum spend limits, sensitive data blocklist, and prompt injection detection — before any Payment Proof is created. An optional Gemini Classifier produces Intent Signals that provide human-readable reasoning for the Policy Decision. Critically, the LLM never overrides deterministic policy, it explains, not decides. Every Payment Attempt, whether allowed or blocked, creates an Audit Event with the full enforcement trace: challenge, decision, reasons, and signing outcome. Blocked attempts expose the x402-shaped Payment Challenge and all block reasons, but no Payment Proof is ever created. The MVP is intentionally simulated, no real wallet signing, facilitator calls, or testnet transactions. Instead, it draws a clear X402 Payment Adapter boundary showing exactly where signing would happen, with inspectable simulated PAYMENT-SIGNATURE and PaymentPayload evidence for allowed attempts. Built with FastAPI (Python), Gemini API for intent classification, and a Guided Enforcement Replay UI, a step-by-step walkthrough of one approved and one blocked canonical Payment Attempt that advances manually so the enforcement boundary is impossible to miss.