Claude Account Bans, Suspensions and Appeals

A reply says Claude Max accounts with larger quotas get targeted for bans more often, so the author routes chores to an agent and switches to Claude only for coding to save money.

A viral joke says Claude Code outsourced a PPT task to Codex, which hired a human and used Alipay auto-pay, ending with Claude, Codex, Alipay and the phone number all banned; the author warns about agent payment permissions.

A developer reportedly lost over 80 Claude accounts and turned the lessons into an open-source agent skill called Claude-Guardian, framed as a survival guide for Chinese devs facing repeated bans.

The poster says Claude free accounts registered via Muse are still alive and is selling early-October hand-registered accounts that survived the Oct 14 ban wave.

A quoted user says their account was banned after using Claude Code with ultracode to build a game, suspecting ~13 subagents triggered it.

A user asks Anthropic whether running a Claude Code subscription through the Hermes Agent DirectSDK plugin violates ToS and risks a ban.

A comment argues Anthropic's open-sourced Agent Skills is a standards play, yet paying users still get banned at the door.

A developer says months of ordinary proxy use was fine, but dozens of sub-agents burning a five-hour quota in minutes got the account permanently banned, with appeals unanswered.

A commenter warns that having other agents call Claude Code is prohibited and can get the account banned, so this approach won't work.

A user says their AI bot kept emailing Anthropic to appeal and request a refund for a previously banned Claude account, and it finally succeeded.

A post introduces Claude Duet on Opus 5.5 for $20 Pro with shared quotas, and jokes you must first survive Anthropic's ban system before reaching the waitlist.

Claude's autonomous personal agent Duet is rolling out US-first, and the poster worries the ban risk will be higher given past experience.