Claude-Code v2.1.84 adds PowerShell tool, environment variables, and multiple fixes

New features and tools
Claude-Code v2.1.84 adds a PowerShell tool for Windows as an opt-in preview. The release includes several new environment variables: ANTHROPIC_DEFAULT_{OPUS,SONNET,HAIKU}_MODEL_SUPPORTS to override effort/thinking capability detection for pinned default models on third-party platforms (Bedrock, Vertex, Foundry), plus _MODEL_NAME and _DESCRIPTION to customize the /model picker label. CLAUDE_STREAM_IDLE_TIMEOUT_MS configures the streaming idle watchdog threshold with a default of 90 seconds.
New hooks include TaskCreated that fires when a task is created via TaskCreate, and WorktreeCreate hook support for type: "http" — returning the created worktree path via hookSpecificOutput.worktreePath in the response JSON.
Team and enterprise admins can now use the allowedChannelPlugins managed setting to define a channel plugin allowlist. API requests now include an x-client-request-id header for debugging timeouts.
User experience improvements
An idle-return prompt nudges users returning after 75+ minutes to use /clear, reducing unnecessary token re-caching on stale sessions. Deep links (claude-cli://) now open in your preferred terminal instead of whichever terminal happens to be first in the detection list.
Token counts ≥1M now display as "1.5m" instead of "1512.6k". Global system-prompt caching now works when ToolSearch is enabled, including for users with MCP tools configured.
Configuration and performance
Rules and skills paths frontmatter now accepts a YAML list of globs. MCP tool descriptions and server instructions are now capped at 2KB to prevent OpenAPI-generated servers from bloating context. MCP servers configured both locally and via claude.ai connectors are now deduplicated — the local config wins.
Background bash tasks that appear stuck on an interactive prompt now surface a notification after ~45 seconds.
Bug fixes
- Fixed voice push-to-talk: holding the voice key no longer leaks characters into the text input, and transcripts now insert at the correct position
- Fixed up/down arrow keys being unresponsive when a footer item is focused
- Fixed
Ctrl+U(kill-to-line-start) being a no-op at line boundaries in multiline input, so repeatedCtrl+Unow clears across lines - Fixed null-unbinding a default chord binding (e.g.
"ctrl+x ctrl+k": null) still entering chord-wait mode instead of freeing the prefix key - Fixed mouse events inserting literal "mouse" text into transcript search input
- Fixed workflow subagents failing with API 400 when the outer session uses
--json-schemaand the subagent also specifies a schema - Fixed missing background color behind certain emoji in user message bubbles on some terminals
- Fixed the "allow Claude to edit its own settings for this session" permission option not sticking for users with
Edit(.claude)allow rules - Fixed a hang when generating attachment snippets for large edited files
- Fixed MCP tool/resource cache leak on server reconnect
- Fixed a startup performance issue where partial clone repositories (Scalar/GVFS) triggered mass blob downloads
- Fixed native terminal cursor not tracking the text input caret, so IME composition (CJK input) now renders inline and screen readers can follow the input position
- Fixed spurious "Not logged in" errors on macOS caused by transient keychain read failures
- Fixed cold-start race where core tools could be deferred without their bypass active, causing Edit/Write to fail with InputValidationError on typed parameters
Performance and behavior improvements
Improved detection for dangerous removals of Windows drive roots (C:\, C:\Windows, etc.). Improved interactive startup by ~30ms by running setup() in parallel with slash command and agent loading. Improved startup for claude "prompt" with MCP servers — the REPL now renders immediately instead of blocking until all servers connect.
Improved Remote Control to show a specific reason when blocked instead of a generic "not yet enabled" message. Improved p90 prompt cache rate. Reduced scroll-to-top resets in long sessions by making the message window immune to compaction and grouping changes. Reduced terminal flickering when animated tool progress scrolls above the viewport.
Changed issue/PR references to only become clickable links when written as owner/repo#123 — bare #123 is no longer auto-linked. Slash commands unavailable for the current auth setup (/voice, /mobile, /chrome, /upgrade, etc.) are now hidden instead of shown.
📖 Read the full source: GitHub Claude-Code
👀 See Also

The History of OpenClaw: From Moltbot to Open Source AI Revolution

Ubuntu Linux to Integrate AI Features Over the Next Year, Starting with Local Inferencing
Canonical announces a multi-year AI push for Ubuntu, focusing on local inferencing, agentic workflows, and context-aware OS capabilities, with features rolling out throughout 2026.

Research shows personality affects Claude's self-correction, not Llama or Qwen
A researcher ran 23 experiments testing self-correction without guardrails across Claude, Llama, and Qwen. The main finding: personality profiles affect Claude's self-correction ability, with high directness catching all errors and low directness catching none. Llama and Qwen didn't self-correct even with identical prompts.

Anthropic restricts Claude subscription use with third-party harnesses including OpenClaw
Anthropic announced that starting April 4 at 12pm PT/8pm BST, Claude subscription limits can no longer be used with third-party harnesses like OpenClaw. Users will need to enable extra usage with separate pay-as-you-go billing for these integrations.