Changelog — Claude Code Cheatsheet

Cheatsheet changes:

+ sandbox.allowAppleEvents → Opt-in: let sandboxed commands send Apple Events on macOS [config-env]

+ CLAUDE_CLIENT_PRESENCE_FILE → Marker file to suppress push notifications while at the machine [config-env]

~ /config: ~~Open settings~~ → Open settings; key=value sets any setting [slash-commands]

- OTEL_RESOURCE_ATTRIBUTES [config-env]

- API_FORCE_IDLE_TIMEOUT [config-env]

Full changelog (39 entries):

• Added `/config key=value` syntax to set any setting from the prompt (e.g. `/config thinking=false`) — works in interactive, `-p`, and Remote Control

• Added `sandbox.allowAppleEvents` opt-in setting that lets sandboxed commands send Apple Events on macOS

• Added `CLAUDE_CLIENT_PRESENCE_FILE` environment variable: point it at a marker file to suppress mobile push notifications while you're at the machine

• Upgraded the bundled Bun runtime to 1.4

• Improved streaming of long paragraphs: text now appears line-by-line instead of waiting for the first line break

• Improved auto-retry: API connection drops mid-thinking now automatically retry instead of showing "Connection closed while thinking"

• Improved the subagent panel: idle subagents auto-hide after 30s, the list caps at 5 rows with scroll hints, and keyboard hints now show in the footer

• Improved the MCP OAuth browser page to match Claude Code's visual style and auto-close on success

• Changed fullscreen mode URL opening to require Cmd+click (macOS) / Ctrl+click, matching native terminal behavior

• Changed the `Improved N memories` line to no longer list individual files outside verbose mode

• Fixed prompt caching not reading on custom `ANTHROPIC_BASE_URL` and on Foundry due to a per-request attestation token changing every turn

• Fixed Write/Edit producing 0-byte or truncated files on network drives and cloud-synced folders

• Fixed `open`, `osascript`, and browser-based auth flows failing with error -600 on macOS by adding the Apple Events entitlement

• Fixed a startup regression (~120ms per launch in fresh environments, introduced in 2.1.169): the first prompt no longer waits for the managed-settings fetch when no MCP servers are configured

• Fixed startup blocking with a blank terminal for up to 15 seconds when the account settings fetch is slow on a degraded network

• Fixed startup crash (`TypeError: Cannot read properties of null`) when `.claude.json` contains corrupted null project entries

• Fixed macOS TUI freezing at session start (Ctrl+C unresponsive) when Spotlight is busy reindexing

• Fixed long-running idle sessions losing their history when another Claude Code process ran the 30-day transcript cleanup

• Fixed foreground subagents spawning unbounded nested chains; they now respect the same 5-level depth limit as background subagents

• Fixed `/recap` and conversation forks using the previous model immediately after a model switch

• Fixed subagent "Thinking" duration showing the parent agent's elapsed time instead of the subagent's own

• Fixed subagents blocked on a nested agent showing a ticking elapsed time instead of "waiting" in the agent panel

• Fixed the API retry indicator ("Retrying in 0s · attempt N/10") staying on screen after the retry succeeded

• Fixed AWS `awsCredentialExport` credentials with a short remaining lifetime causing credential refreshes every minute, and now accepts the JSON shape from `aws configure export-credentials`

• Fixed `claude mcp get`/`list` showing `✓ Connected` when tools/list fails; they now show `! Connected · tools fetch failed` with the error detail

• Fixed `/remote-control` leaving a stale "connecting…" line; it now confirms in the transcript once connected

• Fixed ExitWorktree refusing to remove a clean worktree with "Could not verify worktree state" when bare `git` cannot be resolved on Windows

• Fixed settings changes (such as `/effort` or `/model`) failing with ENOENT when `~/.claude/settings.json` is a relative symlink under a symlinked `~/.claude`

• Fixed IDE selection line numbers in context reminders being off by one (IntelliJ and VS Code)

• Fixed Ctrl+C in fullscreen after a native terminal selection (modifier+drag) overwriting the clipboard with the app's prior selection

• Fixed Ctrl+V showing "No image found in clipboard" instead of pasting when the clipboard contains text

• Fixed agent creation failing with "EEXIST: file already exists" when the agents directory already exists (Windows/OneDrive)

• Fixed AskUserQuestion preview content being cut off at the dialog edge instead of word-wrapping

• Fixed AskUserQuestion multi-select questions silently dropping a typed "Other" free-text answer when submitting

• Fixed `/stats` "Most active day" and daily token chart dates showing one day early in UTC-negative timezones

• Fixed `/copy` and copy-on-select on Linux not detecting a clipboard utility installed after Claude Code started

• Fixed tab-indented code rendering with incorrect indentation in the Write (create-file) preview

• Fixed user prompts queued mid-turn not showing a full-width background highlight in the transcript

• Fixed the activity spinner's pulse dwelling on the wrong glyph size in Ghostty

Cheatsheet changes:

+ Nested .claude/skills → Nested skill dirs load; clashes show as <dir>:<name> [skills-agents]

+ Nested .claude/ priority → Agent/workflow/output-style closest to cwd wins on collision [skills-agents]

~ deny: tool-name glob: * denies all tools; allow rules reject non-MCP globs → * denies all; Tool(param:value) matches input params; allow rejects non-MCP [config-env]

~ workflowKeywordTrigger: ~~Toggle 'ultracode' keyword triggering dynamic workflows~~ → Explicit phrases only ('run a workflow', 'workflow:'); purple shimmer [config-env]

~ Dynamic workflows: ~~Ask Claude to create; orchestrates tens to hundreds of agents (keyword: ultracode)~~ → Orchestrates tens–hundreds of agents; triggers on 'run a workflow', 'workflow:' [workflows-tips]

Cheatsheet changes:

+ language → Pin session title language (overrides auto-detection) [config-env]

+ footerLinksRegexes → Regex-matched link badges in footer row (user or managed) [config-env]

- wheelScrollAccelerationEnabled [config-env]

- terminalSequence [config-env]

Cheatsheet changes:

+ enforceAvailableModels → Managed setting; availableModels constrains Default model, prevents widening [config-env]

Cheatsheet changes:

+ wheelScrollAccelerationEnabled → Disable mouse-wheel scroll acceleration in fullscreen mode [config-env]

- ~/.claude/themes/ [config-env]

- defaultEnabled [config-env]

Cheatsheet changes:

+ Nesting depth → Sub-agents can spawn their own sub-agents up to 5 levels deep [skills-agents]

Cheatsheet changes:

+ --safe-mode → Start with all customizations disabled (CLAUDE.md, plugins, skills, hooks, MCP) [cli-flags]

+ CLAUDE_CODE_SAFE_MODE → Env var equivalent of --safe-mode flag [config-env]

+ /cd [path] → Move session to new working directory without breaking prompt cache [slash-commands]

+ disableBundledSkills → Hide bundled skills, workflows, and built-in slash commands from model [config-env]

+ CLAUDE_CODE_DISABLE_BUNDLED_SKILLS → Env var equivalent of disableBundledSkills setting [config-env]

+ API_FORCE_IDLE_TIMEOUT → Opt out of default 5min idle timeout on Vertex/Foundry (set to 0) [config-env]

~ claude agents --json: ~~List live sessions as JSON; includes waitingFor for blocked sessions~~ → List live sessions as JSON; --all includes completed; id/state fields [cli-flags]

- MessageDisplay [config-env]

- reloadSkills [config-env]

- sessionTitle [config-env]

- WorktreeCreate [config-env]

Cheatsheet changes:

+ fallbackModel → Up to 3 fallback models tried in order when primary is overloaded [config-env]

+ deny: tool-name glob → * denies all tools; allow rules reject non-MCP globs [config-env]

+ --thinking → disabled turns off thinking on models that think by default [cli-flags]

~ --fallback-model: ~~Switch to this model when primary is not found~~ → Fallback when primary unavailable (interactive + headless) [cli-flags]

~ MAX_THINKING_TOKENS: ~~0=off~~ → 0=off; disables thinking on models that think by default [config-env]

- pluginSuggestionMarketplaces [config-env]

- allowAllClaudeAiMcps [config-env]

Cheatsheet changes:

+ /plugin list [--enabled|--disabled] → List installed plugins with status filters [slash-commands]

+ requiredMinimumVersion / requiredMaximumVersion → Managed settings; refuses start if version outside allowed range [config-env]

+ hookSpecificOutput.additionalContext → Stop/SubagentStop: give Claude feedback, continue turn [config-env]

+ \$ → Escape literal $ before digit in command bodies [skills-agents]

~ CLAUDE_CODE_SESSION_ID: ~~Session ID passed to stdio MCP server subprocesses~~ → Session ID passed to stdio MCP server subprocesses (also on --resume) [config-env]

Cheatsheet changes:

~ claude agents --json: ~~List live sessions as JSON for scripting~~ → List live sessions as JSON; includes waitingFor for blocked sessions [cli-flags]

Cheatsheet changes:

+ OTEL_RESOURCE_ATTRIBUTES → Custom dimension labels on OTEL usage metric datapoints [config-env]

Cheatsheet changes:

~ workflowKeywordTrigger: ~~Toggle 'workflow' keyword triggering dynamic workflows~~ → Toggle 'ultracode' keyword triggering dynamic workflows [config-env]

~ Dynamic workflows: ~~Ask Claude to create; orchestrates tens to hundreds of agents~~ → Ask Claude to create; orchestrates tens to hundreds of agents (keyword: ultracode) [workflows-tips]

Cheatsheet changes:

+ CLAUDE_CODE_ENABLE_AUTO_MODE → Enable auto mode on Bedrock/Vertex/Foundry for Opus 4.7/4.8 [config-env]

~ Auto mode: ~~No opt-in consent needed; built-in for Max subscribers~~ → Built-in for Max; Bedrock/Vertex/Foundry via CLAUDE_CODE_ENABLE_AUTO_MODE [workflows-tips]

- OTEL_LOG_TOOL_DETAILS [config-env]

Cheatsheet changes:

+ claude plugin init <name> → Scaffold a new plugin in .claude/skills [cli-flags]

+ workflowKeywordTrigger → Toggle 'workflow' keyword triggering dynamic workflows [config-env]

+ OTEL_LOG_TOOL_DETAILS → Include tool params (bash cmds, MCP/skill names) in OTEL [config-env]

+ EnterWorktree (switch) → Switch between Claude-managed worktrees mid-session [workflows-tips]

~ .claude/skills/<name>/: ~~Project skills~~ → Project skills (auto-loaded as plugins) [skills-agents]

- claude project purge [path] [cli-flags]

- worktree.bgIsolation [config-env]

- CLAUDE_CODE_STOP_HOOK_BLOCK_CAP [config-env]

Cheatsheet changes:

+ ←← → Open agents view [keyboard-shortcuts]

+ /workflows → View dynamic workflow runs [slash-commands]

+ /chrome → Select connected browser for Chrome integration [slash-commands]

+ /logout → Sign out (in agents view) [slash-commands]

+ claude --bg --exec '<cmd>' → Run shell command as attachable background session [cli-flags]

+ defaultEnabled → Plugin default disabled in plugin.json; enable with /plugin [config-env]

+ CLAUDE_CODE_SESSION_ID → Session ID passed to stdio MCP server subprocesses [config-env]

+ /simplify → Cleanup-only review (reuse, simplify, efficiency, altitude) and apply fixes [skills-agents]

+ Dynamic workflows → Ask Claude to create; orchestrates tens to hundreds of agents [workflows-tips]

~ /effort [level]: ~~Set effort; interactive slider (low/med/xhigh/high/max/auto)~~ → Set effort; interactive slider (low–max; Faster/Smarter labels) [slash-commands]

~ Workflow: ~~Deterministic multi-agent orchestration (opt-in)~~ → Dynamic multi-agent orchestration (opt-in); /workflows to view runs [skills-agents]

~ 1M context: ~~Opus 4.7 (Max/Team/Ent)~~ → Opus 4.8 (Max/Team/Ent) [workflows-tips]

~ /effort: ~~○ low · ◐ med · ◑ xhigh · ● high/max~~ → ○ low · ◐ med · ◑ xhigh · ● high/max (Faster↔Smarter) [workflows-tips]

~ claude mcp list: ~~List all servers~~ → List all servers (⏸ = pending approval) [mcp-servers]

- Ctrl+U [keyboard-shortcuts]

- /scroll-speed [slash-commands]

- /color [color] [slash-commands]

- /tui [fullscreen] [slash-commands]

- --channels [cli-flags]

- worktree.baseRef [config-env]

- OTEL_METRICS_INCLUDE_ENTRYPOINT [config-env]

Cheatsheet changes:

~ /model [model]: ~~Switch model (←→ effort, d = set default)~~ → Switch model (←→ effort, s = this session only) [slash-commands]

Cheatsheet changes:

+ /reload-skills → Re-scan skill directories without restarting [slash-commands]

+ disallowed-tools → Remove tools from model while skill is active [skills-agents]

+ pluginSuggestionMarketplaces → Allowlist org marketplaces for plugin suggestions [config-env]

+ reloadSkills → SessionStart: re-scan skills so new ones are available [config-env]

+ sessionTitle → SessionStart: set session title on startup and resume [config-env]

+ MessageDisplay → Transform or hide assistant message text as displayed [config-env]

+ --fallback-model → Switch to this model when primary is not found [cli-flags]

+ / (vim NORMAL) → Reverse history search (like Ctrl+R) [keyboard-shortcuts]

~ /code-review [effort]: ~~Code review with effort level; --comment for inline PR comments~~ → Code review with effort; --comment for inline PR comments, --fix to apply findings [skills-agents]

~ Auto mode: ~~Built-in for Max subscribers~~ → No opt-in consent needed; built-in for Max subscribers [workflows-tips]

- ANTHROPIC_BEDROCK_SERVICE_TIER [config-env]

- DISABLE_UPDATES [config-env]

- PostToolUseFailure [config-env]

- PermissionDenied [config-env]

- PreCompact [config-env]

- --exclude-dynamic-system-prompt-sections [cli-flags]

- Ctrl+E [keyboard-shortcuts]

- /chrome [slash-commands]

Cheatsheet changes:

+ allowAllClaudeAiMcps → Load claude.ai cloud MCP connectors alongside managed-mcp.json [config-env]

~ /usage: ~~Usage stats, cost, and rate status~~ → Usage stats with per-category breakdown, cost, and rate status [slash-commands]

~ /diff: ~~Interactive diff viewer~~ → Interactive diff viewer (keyboard-scrollable detail) [slash-commands]

- CLAUDE_CODE_POWERSHELL_RESPECT_EXECUTION_POLICY [config-env]