Multi-Agent Setup
When you move beyond a single agent, each agent should have its own wallet. This guide covers architecture patterns, budget management, and operational practices for multi-agent deployments.Why Separate Wallets?
Each agent gets its own wallet for four reasons:- Isolation — a compromised or misbehaving agent can only affect its own wallet, not others
- Clear audit trail — every transaction is attributed to a specific agent’s wallet address
- Individual budgets — each agent gets funded according to its role and risk profile
- Independent control — you can pause, refund, or shut down one agent without affecting the rest
Architecture Patterns
Independent Agents
Each agent has its own wallet and operates on its own tasks. No coordination between agents.Team of Agents
Multiple agents serve one workflow, each handling a specific role. Each has its own wallet scoped to its function.Scaled Fleet
Many identical agents doing the same task in parallel, each with their own wallet.Setting Up Multiple Wallets
Each wallet is tied to a District Pass. For multi-agent setups, each agent connects with its own credentials.MCP Configuration
Each agent gets its own MCP connection, authenticated with its own District Pass:FDX CLI Configuration
For agents using the FDX CLI, each authenticates separately:Budget Management
Apply the pocket money philosophy per agent:| Agent Role | Suggested Funding | Refill Strategy |
|---|---|---|
| Research / data buying | $10–50 | Per session |
| API access payment | $50–200 | Weekly |
| Active trading | $500–2,000 | Based on performance |
| Payment processing | Task-appropriate | Per batch |
Monitoring
With each agent having its own wallet address, monitoring is straightforward:- Per-agent activity — look up each wallet address on the relevant block explorer
- Balance tracking — check balances via MCP, CLI, or block explorer
- Spending patterns — review transaction history to verify agents are operating within expectations