Freysa
Adversarial AI agent experiment where users try to convince an agent to release onchain funds.
Overview
Freysa is a series of adversarial agent experiments: an AI agent controls a prize pool onchain and users pay to send messages attempting to persuade it to transfer the funds. It became a notable case study in agent security, prompt-injection resistance, and autonomous custody of funds.
Sato Score
29/100· LowHow open, active & verifiable — not a safety or returns grade.
- Maintenance & liveness
- 20/35
- Code transparency
- 4/20
- Interface verifiability
- 0/10
- Docs & demo
- 0/10
- Listing transparency & provenance
- 5/15
- Independent verification
- 0/10
Details
- Chains
- Base
- Agent types
- Gaming, Security, Social
- Open source
- Partial
- Pricing
- Pay-per-message game
- Last checked
- 2026-07-25
Marketplace Signals
- ▸ Functional agent
Verification status reflects evidence reviewed by Sato Hub. Self-reported claims are never presented as verified.
Where Freysa fits
Related concepts: How Agents Use Wallets, What Are Onchain Agents?, Onchain Agents vs Offchain Agents
Listing History
Recorded changes to Freysa's listing — status, verification, Sato Score, and recent releases. Full history is kept in the listing record.
- Jun 11, 2026Sato Hub began tracking this listing
Related Resources
Turnkey
ActiveSecure key management infrastructure with policy controls, commonly used for agent wallets.
aixbt
ActiveAI market-intelligence agent that publishes crypto narrative and project analysis.
Giza (ARMA)
ActiveARMA is Giza's autonomous agent that manages stablecoin lending positions across Base DeFi protocols on the user's behalf.
Join the Sato Hub Briefing
One email a week — the agents, tools, and infrastructure that actually shipped, and why they matter.