BasePulse is a fully autonomous agent that monitors social trends on X and Farcaster, validates them against on-chain metrics, and autonomously deploys themed ERC20 tokens on the Base blockchain—all without human intervention.

• Autonomous Trend Monitoring: Continuously scans X and Farcaster for Base ecosystem trends using LLM-powered sentiment analysis
• On-Chain Validation: Cross-references social trends with DexScreener metrics to ensure deployment quality
• Autonomous Token Deployment: Deploys ERC20 tokens with Uniswap V4 liquidity pools using Clanker SDK
• Treasury Management: Automatically collects trading fees and reinvests them into Base ecosystem initiatives
• Real-Time Dashboard: Web-based interface for monitoring agent operations, deployed tokens, and treasury
• Community Interaction: Autonomous social media posting and community engagement
• Comprehensive Logging: Full audit trail of all decisions, deployments, and transactions

BasePulse consists of three main components:

Analyzes social media posts using Claude AI to extract trending themes and sentiment scores. Validates trends against on-chain volume data before triggering deployments.

Handles autonomous ERC20 token deployment on Base using the Clanker SDK. Manages wallet interactions, transaction execution, and treasury accounting.

Coordinates the entire workflow, running at regular intervals to scan trends, validate them, and trigger deployments. Includes safety limits and error handling.

BasePulse implements several key blockchain primitives:

• ERC20 Token Creation: Deploy custom tokens with metadata and social links
• Uniswap V4 Integration: Create liquidity pools with anti-sniper protection
• Treasury Management: On-chain fund tracking and reinvestment logic
• Fee Collection: Automatic fee harvesting from deployed tokens
• Transaction Tracking: Full audit trail of all on-chain operations

Unlike traditional chatbots, BasePulse acts as a venture agent that:

1. Identifies promising trends in the Base ecosystem
2. Deploys capital (tokens) to capture those opportunities
3. Manages the resulting treasury autonomously
4. Reinvests profits into new initiatives

This creates a self-sustaining economic loop where the agent generates value without human direction.

• Frontend: React 19 + Tailwind CSS 4 + shadcn/ui
• Backend: Express 4 + tRPC 11
• Database: MySQL with Drizzle ORM
• Blockchain: Viem + Clanker SDK for Base network
• AI/LLM: OpenAI Claude for trend analysis
• Agent Framework: OpenClaw for autonomous operations

• Node.js 22+
• pnpm
• MySQL database
• Base network RPC endpoint
• Private key with ETH for gas
• OpenAI API key

# Clone repository
git clone https://github.com/basepulse/agent.git
cd basepulse-agent

# Install dependencies
pnpm install

# Configure environment
cp .env.example .env.local
# Edit .env.local with your configuration

# Set up database
pnpm db:push

# Start development server
pnpm dev

# Development
pnpm dev

# Production
pnpm build
pnpm start

Open http://localhost:3000/dashboard to monitor:

• Deployed tokens and their performance
• Treasury balance and transaction history
• Real-time trend analysis and sentiment scores
• Community interactions and agent responses

Edit the configuration in server/agent/autonomousLoop.ts:

const config = {
  enabled: true,
  intervalMinutes: 15,              // Run every 15 minutes
  minSentimentScore: 60,            // 0-100 scale
  minMentions: 5,                   // Minimum mentions
  minVolume24hUSD: 100000,          // Minimum on-chain volume
  maxDeploymentsPerDay: 10,         // Safety limit
};

• Agent Architecture - Detailed technical architecture and design
• Deployment Guide - Production deployment and configuration
• API Reference - tRPC API endpoints

BasePulse uses the following main tables:

• deployed_tokens: Tracks all tokens deployed by the agent
• trend_analysis: Historical trend analysis results
• treasury_transactions: All treasury movements
• social_interactions: Community engagement records
• agent_metrics: Performance and operational metrics

• Private keys stored securely in environment variables
• Rate limiting on deployments (max 10 per day)
• Multiple validation checks before deployment
• Comprehensive error handling and logging
• Owner notifications for all significant events

The dashboard displays:

• Total tokens deployed
• Treasury balance and growth
• Deployment success rate
• Average sentiment score of deployed tokens
• Community engagement metrics

1. Scan - Fetch recent posts from X/Farcaster
2. Analyze - Extract trends and sentiment using LLM
3. Validate - Check on-chain metrics and thresholds
4. Deploy - Create token if conditions met
5. Collect - Harvest trading fees
6. Reinvest - Allocate treasury funds
7. Report - Log metrics and notify owner

View all deployed tokens with:

• Token name, symbol, and trend theme
• Sentiment score and market cap
• 24h volume and holder count
• Current status (pending, deployed, active, inactive)

Monitor financial health:

• Total available balance
• Fee collection history
• Reinvestment tracking
• Transaction details

Track trending themes:

• Real-time sentiment scores
• Mention counts
• On-chain volume validation
• Deployment triggers

Engage with users:

• Recent social media posts
• Sentiment analysis (positive/neutral/negative)
• Agent responses
• Engagement metrics

This agent was built for the Base Build OpenClaw Agent Competition with a 5ETH prize pool. It demonstrates:

✅ Implementation of On-Chain Primitives

• Token deployment with Uniswap V4
• Treasury management on-chain
• Fee collection and reinvestment

✅ Novelty of Use Case

• Autonomous venture agent model
• Trend-to-token pipeline
• Self-sustaining economic loop

✅ No-Human-In-The-Loop

• Fully autonomous decision-making
• Autonomous social posting
• Autonomous fund management

The development process included:

1. Research on OpenClaw framework and Base ecosystem
2. Design of autonomous agent architecture
3. Implementation of trend analysis engine
4. Integration of Clanker SDK for token deployment
5. Dashboard development for monitoring
6. Comprehensive testing and validation

See AGENT_ARCHITECTURE.md for detailed technical documentation.

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Submit a pull request

MIT License - See LICENSE file for details

• GitHub: https://github.com/basepulse/agent
• Base Network: https://base.org
• Clanker: https://clanker.world
• OpenClaw: https://openclaw.ai

For questions or issues:

• Check the Deployment Guide
• Review Agent Architecture
• Open an issue on GitHub

BasePulse - Autonomous Trend-to-Token Engine for Base Ecosystem

Built with ❤️ for the Base Build OpenClaw Agent Competition