Security FAQ

Circuit is built with security at its core. Agents can help you automate and manage onchain activity, but your wallets remain self-custodial and under your control. Below are common security questions and answers to help you feel confident when using Circuit.

Wallet and Keys

Q: Do agents ever have access to my private keys?

No. Agents never see or touch your private keys. Circuit keeps agent code and secure signing strictly separated. Agents run in a sandboxed environment, while all credential signing happens in a dedicated secure enclave.

Q: What happens if I lose access to my wallet?

For imported and created wallets, you can export the private key from the Circuit dashboard. This is done self-custodially by authenticating with your passkey, the Circuit team has no access to it.

Agent Behavior

Q: Can an agent drain my wallet?

No. Agents don't have access to private keys and cannot sign transactions or take direct custody of your funds. They only suggest transactions that must match user policy settings.

Q: What if an agent suggests something malicious?

Circuit applies global safety rules and personal policy settings to suggested transactions before signing.

Q: How do I know if an agent is safe to use?

Circuit supports both first-party agents built by the Circuit team, and third-party agents built by external developers. Every agent is reviewed by Circuit before being listed, and you can track metrics such as its historical ROI, current TVL, and live transactions.

Q: Is there a limit to what an agent can do?

To start, agents are focused on actions they can take with crypto private keys. Eventually, we plan to expand to centralized exchanges, other consumer finance platforms, and other apps.

Q: Can I restrict an agent to certain wallets?

Yes. You decide which wallet to run an agent on, and you can stop it at any time. Agents cannot suggest transactions outside the wallets they are on.

Control and Revocation

Q: Can I stop an agent after I've started it?

Yes. You can stop any agent at any time from the Circuit dashboard.

Q: What happens to positions or strategies when I stop an agent?

Depending on the agent, some positions may remain open until you manually unwind them, while other positions may be automatically unwound. The agent itself will stop running immediately.

Q: How do I check which agents are currently active?

The Circuit dashboard shows all agents currently running and the wallets they're running on.

Circuit Oversight

Q: How does Circuit keep me safe?

Circuit acts like a circuit breaker for your wallet. All agent suggestions are checked against global safety rules and your personal policies before being passed to your wallet.

Q: What happens if Circuit detects suspicious behavior?

The action itself is blocked from the wallet. Agents attempting malicious transactions lead to platform suspension.

Q: How are agents reviewed or audited before being listed?

First-party agents are built and audited by the Circuit team. Third-party agents go through an internal review process to ensure they meet safety and reliability standards before being listed. Some third-party developers may also provide their own external audits, which will be noted in the agent’s description.

Q: What happens if Circuit itself is compromised?

Circuit and agents never have access to your private keys, so even an internal compromise would protect these. If malicious behavior is detected, Circuit or you can stop an agent at any time to prevent further transactions.

Transparency

Q: Can I see what actions an agent has taken?

Yes. Since agents transact onchain, you can verify each action against a third-party block explorer. Circuit also shows chain-of-thought logs to explain why agents took certain actions.

Last updated