Descriptions:
Zubair Trabzada of AI Workshop demonstrates a multi-agent crypto research system he built with Claude Code that generates comprehensive investment research reports on any token by deploying five AI agents in parallel. Orchestrated through a skills.md file, the system runs simultaneous agents for on-chain wallet and exchange flow analysis, tokenomics and unlock schedules, sentiment aggregation across Twitter and Reddit, technical charting, and fundamental assessment — then synthesizes all five outputs into a scored markdown or PDF report.
In a live walkthrough, Trabzada runs the tool on Solana, producing a grade-B analysis (65/100) that flags $9 billion in TVL at an all-time high, roughly 60,000 new SOL minted daily from ~4% annual inflation, a head-and-shoulders chart pattern with support at $79, and a May 10th token unlock as a near-term supply risk — complete with entry zones ($75–$79), price targets ($130–$145), and a stop-loss at $70. He then runs it live on Ethereum, explaining each of the three execution phases: discovery, parallel agent deployment, and synthesis into a final report.
The installation walkthrough covers setting up Claude Code inside VS Code, cloning the repository, and invoking slash commands including a PDF report generator. Trabzada frames the project as a reusable template for building domain-specific multi-agent research workflows using skills files — applicable beyond crypto to any area requiring parallel specialist analysis.
📺 Source: Zubair Trabzada | AI Workshop · Published May 02, 2026
🏷️ Format: Hands On Build
