Genuinely curious: is anyone else doing blockchain work with Claude, and how have the security measures been affecting your work?
A Reddit user tried to use Claude for blockchain work; its security classifiers flagged about 8 legitimate prompts in 3 days, blocking productivity and leaving a refund denied.
Summary written by AI. Read the source before acting on anything.
Built this, or have results since? Suggest an update. See more on real crypto AI agent use cases.