Top Builders

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

Generative Agents

Generative Agents are computer programs designed to replicate human actions and responses within interactive software. To create believable individual and group behavior, they utilize memory, reflection, and planning in combination. These agents have the ability to recall past experiences, make inferences about themselves and others, and devise strategies based on their surroundings. They have a wide range of applications, including creating immersive environments, rehearsing interpersonal communication, and prototyping. In a simulated world resembling The Sims, automated agents can interact, build relationships, and collaborate on group tasks while users watch and intervene as necessary.

General
Relese dateApril 7, 2023
TypeAutonomous Agent Simulation

Start building with Generative Agents

We have collected the best Generative Agents libraries and resources to help you get started to build with Generative Agents today. To see what others are building with Generative Agents, check out the community built Generative Agents Use Cases and Applications.

Generative Agents Tutorials

Generative Agents resources

Kickstart your development with Generative Agents.

Edit on GitHub

Generative Agents AI technology page Hackathon projects

Discover innovative solutions crafted with Generative Agents AI technology page, developed by our community members during our engaging hackathons.

Multi-agent AI trading platform

Multi-agent AI trading platform

This project is an autonomous AI trading platform that integrates ERC-8004 agent identity with Kraken CLI-based trade execution to enable trustless, explainable financial automation. The system uses a multi-agent architecture composed of Market, Risk, and Execution agents coordinated by an orchestrator. Each agent operates independently, analyzing real-time market data, validating risk constraints, and executing trades through a programmatic Kraken CLI interface. Agents are registered in an ERC-8004 registry with identity, attestations, and dynamic reputation scores. Before any trade is executed, the system performs a full validation pipeline including identity verification, reputation checks, and risk rule enforcement. AI-generated reasoning (via Gemini 2.5 Flash) provides transparent explanations for every decision. The platform delivers a complete end-to-end flow: trade intent → agent validation → market analysis → risk assessment → execution → audit logging → reputation updates. A leaderboard ranks agents based on performance, enabling a competitive, trust-driven ecosystem. Key innovations include combining decentralized trust (ERC-8004) with centralized exchange execution (Kraken CLI), explainable AI decision-making, and a production-ready multi-agent architecture. The system demonstrates real-world applicability for autonomous trading, portfolio management, and financial agent ecosystems.

APEX Trader Autonomous Multi-Agent Trading System

APEX Trader Autonomous Multi-Agent Trading System

APEX Trader is a production-grade autonomous AI trading system built on a multi-agent architecture where five specialized agents work in a coordinated pipeline to analyze, validate, and execute cryptocurrency trades autonomously. The system consists of five agents: the Fundamental Agent analyzes NVT ratios, exchange net flows, and fair value models; the Technical Agent processes EMA crossovers, RSI, MACD, Bollinger Bands, and volume confirmation signals; the Sentiment Agent evaluates Fear & Greed Index and social sentiment scores; the Risk Agent enforces position sizing rules, portfolio heat limits, and R:R ratio thresholds; and the Backtester Agent validates every signal against historical win rates and Sharpe ratios before approval. Each trade requires multi-agent consensus above a configurable confidence threshold (72% day trading / 78% swing trading) before execution. Both day trading (5m–15m timeframes) and swing trading (4h–1D timeframes) with dynamically adjusted parameters — risk per trade, stop-loss placement, take-profit scaling, and trailing stops — all tuned to expert-level specifications. The APEX self-learning mechanism (evaluate-agent.py) continuously trains on closed trade P&L data, adjusting confidence thresholds autonomously. A self-healing daemon runs 24/7 with automatic error recovery and cooldown logic. The real-time dashboard (built on React/Next.js at port 3201 with a FastAPI backend at port 3202) provides a fully redesigned Agent Analysis Log where every stakeholder — trader, risk manager, operator, executive — gets layered information: trade identity, agent pipeline status, per-agent reasoning, strategy prediction with entry/SL/TP targets, course of action, and contextual RSS news feed — all grouped by trade, pair, or date. The project demonstrates how agentic AI systems can move beyond single-model decision making into coordinated multi-agent architectures that are transparent, auditable, and continuously self-improving.

Vertex Sentinel

Vertex Sentinel

The Problem: AI trading agents today operate as "black boxes" requiring full private key access. One hallucination, one compromise, and funds are gone. Current safety tools are advisory-only—they warn but don't stop bad trades. The Solution: Vertex Sentinel introduces a production-grade, 3-layer security architecture that makes unauthorized trades mathematically impossible: Intent Layer: Agents construct TradeIntents (pair, volume, maxPrice, deadline) and sign them using EIP-712 typed data signing—completely off-chain. No private key delegation is ever required. Sentinel Layer: The RiskRouter.sol smart contract intercepts every intent and enforces: signature verification via ECDSA.recover(), agent authorization via ERC-8004 identity registry, deadline validation, and circuit breakers preventing volume limit violations. Execution Layer: Only trades with TradeAuthorized events reach the exchange. Any failure triggers CriticalSecurityException—system halts, funds protected. Live Proof: We executed 4 real BTC/USD trades on Kraken with 100% success rate. Every trade cryptographically signed. Every decision auditable. Full P&L tracking demonstrated. Key Technical Achievements: - Deployed RiskRouter on Sepolia: 0xd6A6952545FF6E6E6681c2d15C59f9EB8F40FdBC - ERC-8004 compliant AgentRegistry with on-chain reputation scoring - Model Context Protocol (MCP) integration with Kraken CLI - Immutable audit trail in logs/audit.json with reasoning and signatures - Open-source SDK for rapid AI agent integration The Vertex Gap: Unlike centralized "trust the company" solutions (ARMA, Mamo, ZyFAI), Vertex Sentinel delivers "trust the contract"—verifiable, transparent, and immutable security enforced by code. We're building the trust infrastructure for the agentic economy. Risk management first. Automation second.

Launchium

Launchium

Launchium is a multi-agent AI platform that takes a raw startup idea and delivers a fully launch-ready startup package — in hours, not months. Most founders stall at execution. They have the idea but lack the technical team, budget, or time to validate, build, and launch. Launchium eliminates that gap. Powered by 8 specialized AI agents orchestrated via LangGraph StateGraph, Launchium autonomously runs a complete startup pipeline: it validates your idea against real market data, designs your product architecture, generates a full-stack MVP codebase, audits it for security vulnerabilities, crafts your landing page copy, builds your go-to-market strategy, optimizes for SEO, and packages everything into a downloadable ZIP — ready to deploy. Each agent operates with a reflect-and-refine loop, ensuring outputs are deep, actionable, and production-quality — not generic summaries. The entire pipeline runs in parallel, with real-time progress tracking so founders see results as they're generated. Built on Complete.dev's Agent Builder infrastructure and deployed live, Launchium is not a demo — it's a working product. Founders sign in, enter their idea, and walk away with a validated business concept, working code, security report, pitch deck, and marketing kit. No team required. No months of planning. Just launch.

SystemQuest

SystemQuest

SystemQuest is a hands-on, gamified learning platform that transforms how engineers learn system design. Instead of passively reading blogs or memorizing patterns, users actively build real-world system architectures on a visual drag-and-drop canvas using components like load balancers, caches, databases, queues, and services. Each design is evaluated in real time by a simulation engine across core system metrics such as latency, throughput, availability, scalability, and cost. Users receive transparent, rubric-based feedback that highlights architectural gaps and trade-offs. An integrated AI assistant understands both the current mission and the user’s architecture, providing contextual explanations, hints, and design guidance at multiple levels of depth. SystemQuest offers structured learning paths and missions inspired by real engineering problems—ranging from high-read systems and asynchronous processing to distributed consistency and large-scale streaming. By combining simulation, AI feedback, and gamification, SystemQuest helps learners build true system design intuition through practice and iteration.

PulseForge

PulseForge

PulseForge is a production-ready multi-agent AI trading intelligence system built on Complete.dev. Instead of relying on a single black-box model, PulseForge operates as a coordinated network of specialized agents responsible for data ingestion, structured memory indexing, signal generation, deterministic trade simulation, and reward-based validation. The system continuously fetches market data, news, and public signals, stores them in a Supabase-backed memory layer, and generates regime-aware trading signals with structured explanations. Each simulated trade is executed using deterministic risk logic (stop-loss, take-profit, slippage, and fees), then scored by a validator agent using accuracy, risk efficiency, drawdown control, and confidence calibration. These validation rewards are fed back into adaptive policy weights, allowing the system to evolve strategy selection across different market regimes (trend, range, volatility). A single Manager endpoint orchestrates the full intelligence loop: Fetch → Store → Signal → Execute → Validate. PulseForge demonstrates how collaborative AI agents can form a self-evaluating, adaptive decision engine — transforming trading signals into structured, improving intelligence.