Agents
Claude Code Adds Fallback Models, Cross-Session Message Security, and Broader Deny-Rule Globs
Anthropic's early-June Claude Code updates add fallback model support, broader deny-rule glob matching, stronger cross-session message security, and more reliable thinking controls, alongside improved retries, agents filtering, and a range of terminal/auth/session/UI fixes. The cross-session message hardening and deny-rule globs specifically target safer autonomous and multi-agent runs. Incremental but relevant for anyone running Claude Code unattended in pipelines.
↳ Follow the thread