claude agents gets massive stability pass: session restore, frozen UIs, and history loss fixed; security guardrails added for shell writes; workflow renamed to ultracode.
.zshenv, .zlogin, .bash_login) and ~/.config/git/, which could otherwise lead to unintended command executionclaude agentsclaude rm/stop, idle reap) to send SIGTERM to running shell subprocesses before SIGKILL, so cleanup handlers runclaude agents dropping chat history and re-running the original promptclaude --bg occasionally failing with "socket missing" when the background daemon was cold-starting on a loaded machineclaude rm until the background daemon exitedclaude agents freezing for several seconds when returning to the session list due to the auto-updater re-checking on every exitclaude agents viewfile:///C:/... links being rewritten to a broken path on Windows terminals with hyperlink supportCLAUDE_CODE_ENABLE_AUTO_MODE opt-in instead of incorrectly blaming the model--model when running via the SDK or other hosts where the CLI flag doesn't applyp pasting on the line below instead of at the cursor when the register was yanked with v$grep: single-file grep/egrep/fgrep commands now satisfy the read-before-edit check/effort ultracode incorrectly blaming the dynamic workflows setting when the model cannot run xhigh; ultracode is no longer offered on models that do not support itCLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE; the environment variable is now a no-opworkflow to ultracode. The word "workflow" no longer triggers a run; asking for one in your own words still works. The trigger keyword is highlighted in violet in the prompt inputacceptEdits mode now prompts before writing build-tool config files that grant code execution (.npmrc, .yarnrc*, bunfig.toml, .bazelrc, .pre-commit-config.yaml, .devcontainer/, etc.)