Claude Haiku 5.5 debuts as the default Haiku model (1M context, cheaper pricing), amid a huge fix wave: MCP memory leak, Chrome /tui disconnects, reverted behaviors.
claude-haiku-5-5), now the default Haiku model on the Anthropic API — 1M context, $0.10/$0.50 per Mtok ($0.50/$2.50 for prompts over 100K)agentType to the subagentStatusLine payload, so scripts can tell custom subagent types apartisDeferred to $.tool.register for mods: false lists the tool's schema in the prompt from the start instead of behind tool search← moved the session to the background; if a queued message can't move, ← now stays put and says so/model effort ←/→ wrapping past the highest or lowest level, which could accidentally save Low as a model's default effort/tui disconnecting Claude in Chrome in a session started with --chrome, and ignoring --no-chromeSendMessage in sessions, including resumed ones, where a host, a permission rule or a --tools list removed that tool--agent sessions being told a built-in tool was disabled for the whole session when only their own tool list left it out/clearclaude logs, stop, kill, rm and claude daemon status, stop, uninstall sometimes signing you out when your login had expired or was about toworker being shown as "Agent" when it starts, and the agent detail dialog's title losing the agent type once the agent finishesArtifact("(unprintable path)") while a publish call was still streaming in/ultrareview upload on Linux wrongly refusing some repositories, such as one inside another checkout, over a settings file that "could not be parsed" while a sandboxed command was running/ultrareview upload's refusal over split-index files advising a git command that could leave git unable to read its indexclaude remote-controlclassic.* events being skipped while the plugin hooks worker restarts, which left settings hooks to answer without themclaude plugin test failing for mods that call $.session.append; tests can read the appended rows back with the new mock.sessionclaude plugin eval refusing every Bash-granting run on Macs with Docker Desktop (links under ~/.docker/bin); the refusal now names which part of a credential store held the linkdesktop policy sets Claude Desktop's built-in browser keys, such as builtinBrowserEnabledclaude agents offering bypass permissions that background sessions then ignored when consent was saved only in .claude/settings.local.json or a --settings file; it now asks for consent first, and a session that ignores bypass shows a short notice that stays← backgrounding a session after 10 seconds while the prompt held unsent text (it now cancels the move) or a question was waiting for your answer← had just been pressed to move the session to the background/feedback returning to the drafts list after Ctrl+O or Ctrl+Z while a report was sending, leaving the send impossible to cancelclaude purge stopping silently (exit 0, or a hang in a terminal) when a file or folder could not be deleted; it now deletes the rest, lists what it could not delete, and exits 1>> and << on a line of only spaces leaving the cursor past the end of the line, where a following x deleted nothingV then d) the cursor lands on the first non-blank, and . after it acts on the cursor's lineclaude_code.at_mention logging to emit at most 100 agent and 100 MCP-resource events each time a prompt is read/loop wakeup or scheduled task; Claude is no longer told, and the session stays asleep