Changelog

getting-started: correct the system RAM and hardware requirement (EN + AR)

src/content/modules/getting-started.md and ar/getting-started.md claimed 'at least 4 GB of RAM — 8 GB is recommended for comfortable use'. The Advanced setup page documents hardware as '4 GB+ RAM, x64 or ARM64 processor' and makes no 8 GB recommendation, so the unsupported figure was removed and replaced with the documented requirement: 'at least 4 GB of RAM and an x64 or ARM64 processor'. Verified against https://code.claude.com/docs/en/setup.

memory: document AGENTS.md as project instructions (EN + AR)

Added an AGENTS.md Support section to src/content/modules/memory.md and ar/memory.md, a new quiz question q6 (src/data/modules/memory/quiz.yaml + ar), a feature-catalog entry memory-agents-md (src/data/feature-catalog.json + ar), and a reference-cheatsheet row (src/data/reference-cheatsheet.yaml + ar). Since Claude Code v2.1.277 Claude reads AGENTS.md as project instructions; the default (claude-md-or-agents-md) reads it only when there is no CLAUDE.md, .claude/CLAUDE.md, or CLAUDE.local.md in the working directory or above it, and Project instructions in /config (or the agents-md@builtin plugin's instructionFiles option) changes the behavior. Verified against https://code.claude.com/docs/en/memory.

#560

advanced-features: document the v2.1.278 auto mode server and /status row (EN + AR)

Added a paragraph to src/content/modules/advanced-features.md and ar/advanced-features.md and two feature-catalog entries (claude-code-auto-mode-server and auto-mode-server-status-row, EN + AR). On Enterprise plans, Claude API accounts, and the AWS, Bedrock, Vertex, and Foundry platforms (subject to rollout), auto mode asks the server to run its classifier safety checks as part of the session's model requests with no per-request charge; a gateway or proxy that cannot pass the checks through makes the session fall back to Claude Code's own billed requests, shown as Disabled in the /status Auto mode server row. CLAUDE_CODE_AUTO_MODE_SERVER=0 forces the fallback; it is not read on a direct Anthropic API connection and is temporary. Verified against https://code.claude.com/docs/en/auto-mode-classifier-billing and https://code.claude.com/docs/en/auto-mode-config.

#562

commands: remove the unsupported CLAUDE_CODE_FILE_READ_MAX_OUTPUT_TOKENS tuning claim (EN + AR)

The Read partial-view content in src/content/modules/commands.md (+ ar), the commands quiz (src/data/modules/commands/quiz.yaml + ar), and the feature catalog (src/data/feature-catalog.json + ar) claimed a per-read token budget tunable via CLAUDE_CODE_FILE_READ_MAX_OUTPUT_TOKENS. The tools-reference documents the PARTIAL view and offset/limit pagination but no such env var or setting, so the tuning claim was removed and replaced with the documented behavior (paginate, or narrow with grep; an over-large explicit limit returns an error asking for a smaller limit). Verified against https://code.claude.com/docs/en/tools-reference.

hooks: correct the WorktreeCreate description (EN + AR)

src/content/modules/hooks.md (+ ar) described WorktreeCreate as a v2.1.84 event supporting type http for remote notifications. Per the hooks reference the event (added v2.1.50) fires when a worktree is being created via --worktree, isolation worktree, or for a background session and replaces the default git operation; exit code 2 is not honored and any non-zero exit aborts creation. Rewrote the sentence to match. Verified against https://code.claude.com/docs/en/hooks.

plugins: correct the monitors declaration forms and clarify plugin naming (EN + AR)

src/content/modules/plugins.md (+ ar) claimed the top-level monitors manifest form warns and a future release would require the nested form. Per the docs, monitors are declared in monitors/monitors.json at the plugin root or inline under experimental.monitors; rewrote the paragraph accordingly. Also clarified that an omitted manifest derives the plugin name from the directory (a cache-copied plugin's directory name is a version string), so an explicit name field is recommended for stable identity — matching the plugins quiz. Verified against https://code.claude.com/docs/en/plugins-reference.

plugins: document /plugin install --marketplace one-step add-and-install (EN + AR)

Added the v2.1.275 --marketplace flag to src/content/modules/plugins.md and src/content/modules/ar/plugins.md (install code block plus an explanatory paragraph), the reference cheatsheet (src/data/reference-cheatsheet.yaml + ar), the feature catalog (src/data/feature-catalog.json + ar, id plugin-install-marketplace), the terminal simulator /plugin output (src/data/terminal-commands.json + ar), and a new plugins quiz question q12 (src/data/modules/plugins/quiz.yaml + ar). The flag adds a marketplace and installs a plugin in one step: the source (owner/repo, git URL, or local path, no spaces) is named with --marketplace, the plugin name is given bare, and an unregistered source is confirmed before it's added. Verified against https://code.claude.com/docs/en/discover-plugins (Add a marketplace and install in one command).

plugins: replace the undocumented github: install form in the cheatsheet and terminal output (EN + AR)

The v1.27.5 fix removed '/plugin install github:user/repo' from the plugins module markdown but missed two other surfaces: the reference cheatsheet (src/data/reference-cheatsheet.yaml + ar) and the terminal simulator's /plugin help output (src/data/terminal-commands.json + ar). Both still presented the github: prefix as the way to install directly from a GitHub repo, which the docs don't document. Replaced them with the documented --marketplace one-step flow. Verified against https://code.claude.com/docs/en/discover-plugins.

workflows: correct the --host-config-snapshot disk/memory semantics (EN + AR)

The v1.27.6 copy attributed the 64 MiB limit to disk mode and framed disk as the option that avoids silently dropping large host config; the official docs put the 64 MiB heap cap on memory mode (over which sessions start without host config and show a notice) and describe disk as copying the snapshot under --base-dir with per-file in-memory digest verification at session start (a modified file fails the session and the runner refuses further sessions until restart), with a failed disk write falling back to memory. Rewrote the paragraph in src/content/modules/workflows.md and src/content/modules/ar/workflows.md to match. Verified against https://code.claude.com/docs/en/self-hosted-environments-reference.

#550

workflows: document --drain-marker-file and SELF_HOSTED_RUNNER_DRAIN_MARKER_FILE (EN + AR)

Added one short paragraph to the Self-Hosted Runner section of src/content/modules/workflows.md and src/content/modules/ar/workflows.md covering the v2.1.271 --drain-marker-file <path> flag and its matching SELF_HOSTED_RUNNER_DRAIN_MARKER_FILE env var. The flag makes the runner report its exit to the server as a host drain (telemetry only) when that file exists at a SIGTERM drain; the env var sets the same from a systemd unit or container entrypoint. Verified against https://raw.githubusercontent.com/marckrenn/claude-code-changelog/v2.1.271/meta/cli-surface.md.

workflows: document --host-config-snapshot and SELF_HOSTED_RUNNER_HOST_CONFIG_SNAPSHOT (EN + AR)

Added one short paragraph to the Self-Hosted Runner section of src/content/modules/workflows.md and src/content/modules/ar/workflows.md covering the v2.1.271 --host-config-snapshot disk|memory flag and its matching SELF_HOSTED_RUNNER_HOST_CONFIG_SNAPSHOT env var. The flag chooses how the runner snapshots host config (settings, skills, plugins, MCP servers) at session start; default disk avoids the silent loss when the host config directory exceeds 64 MiB, and memory keeps the snapshot in process with no on-disk cache. Verified against https://raw.githubusercontent.com/marckrenn/claude-code-changelog/v2.1.271/meta/cli-surface.md.

plugins: replace the github: install prefix with the documented owner/repo + @marketplace flow (EN + AR)

Replaced '/plugin install github:username/my-plugin' in src/content/modules/plugins.md (line 161) and src/content/modules/ar/plugins.md (line 162) with the documented two-step flow — '/plugin marketplace add owner/repo-name' followed by '/plugin install plugin-name@marketplace-name' — because the discover-plugins docs only document the plain owner/repo shorthand for the marketplace add step and never use a github: prefix anywhere in the v2.1.270 docs. Verified against https://code.claude.com/docs/en/discover-plugins.md#add-from-github (Add a GitHub repository that contains a `.claude-plugin/marketplace.json` file using the `owner/repo` format, where `owner` is the GitHub username or organization and `repo` is the repository name).

changelog: route every read through one schema-validated loader (EN + AR)

Added loadChangelog(locale) in src/lib/changelog.ts as the single way in: it reads the locale's YAML, parses it, and validates it against a new strict zod schema in src/lib/schemas.ts (non-empty releases; vMAJOR.MINOR.PATCH and unique versions; real-calendar YYYY-MM-DD dates; non-empty titles/summaries; at least one entry; positive-integer references; unknown keys rejected, not stripped). The English page, the Arabic page, and the search-index builder all call it — no production path parses changelog YAML independently any more. A validation failure is re-raised naming the offending release version and field instead of a bare positional path.

#505, #509

changelog: derive the category union and render order from one constant

Replaced the two independent category lists (the type union and the render-order array) with a single exported CHANGELOG_CATEGORIES constant. The schema's enum, the ChangelogCategory type, and groupEntriesByCategory's render order all derive from it, so adding a category is one edit in one place and a schema-valid category can no longer be silently dropped at render time — the mechanism behind the original orphaned-note bug.

#508

changelog: fail the build when the two locales' data drift

Added a pure assertChangelogLocaleParity(en, ar) that throws, naming the versions that disagree, when the two locales differ on the set of release versions, on a shared version's date, or on a shared version's entry count. It runs once per build from the search-index endpoint's getStaticPaths (not as a module-side effect) and from the test suite. It deliberately does not compare categories or human-readable text; the documented limitation is that equal entry counts cannot prove a substituted entry.

#510

changelog: delete the unused feedback-driven flag

Removed the feedbackDriven flag from the changelog entry type, the badge markup on both the English and Arabic pages, and the changelog.feedbackDriven string in both locales. It was defined end to end but set on zero of 333 entries, so it had never rendered anything; deleting it removes dead markup and a misleading field, with no change to rendered output.

#507

advanced-features: document the OTEL_METRICS_INCLUDE_REPOSITORY telemetry variable (EN + AR)

Added OTEL_METRICS_INCLUDE_REPOSITORY to src/content/modules/advanced-features.md and the Arabic counterpart, in the OpenTelemetry environment-variable section beside OTEL_RESOURCE_ATTRIBUTES, with a runnable example. Set to true (default false, requires Claude Code v2.1.269+) it tags every metric and event with vcs.repository.url.full, vcs.owner.name, vcs.repository.name and vcs.provider.name derived once per session from the origin remote, and commit events additionally carry vcs.ref.head.* when OTEL_LOG_TOOL_DETAILS is set. Verified against https://code.claude.com/docs/en/monitoring-usage and the Claude Code v2.1.269 changelog.

#545

workflows: document --remove-session-state behavior, not just the variable name (EN + AR)

Replaced the one-line SELF_HOSTED_RUNNER_REMOVE_SESSION_STATE mention in src/content/modules/workflows.md and the Arabic counterpart with the CLI flag --remove-session-state (default off), its effect (deletes each session's per-session directories under <base-dir>/_sessions/ on session end), and the paired env var, plus a runnable example. Verified against the Claude Code v2.1.268 changelog entry: 'Added claude self-hosted-runner --remove-session-state (default off): delete each session's per-session directories under <base-dir>/_sessions/ when the session ends.'

#542

workflows: name the SELF_HOSTED_RUNNER_REMOVE_SESSION_STATE variable (EN + AR)

Added SELF_HOSTED_RUNNER_REMOVE_SESSION_STATE to src/content/modules/workflows.md (Self-Hosted Runner section) and the Arabic counterpart, as a sibling mention to the existing SELF_HOSTED_RUNNER_CLIENT_LABEL paragraph. The variable ships in Claude Code v2.1.268. Verified against the pinned CLI surface at https://raw.githubusercontent.com/marckrenn/claude-code-changelog/v2.1.268/meta/cli-surface.md.

subagents: scope the --system-prompt-snapshot provider caveat to before v2.1.268 (EN + AR)

Updated src/content/modules/subagents.md and the Arabic counterpart so the note that Bedrock, Google's Cloud Agent Platform, and Foundry rebuild the system prompt on every request and ignore --system-prompt-snapshot reads as a past limitation ('Before v2.1.268, sessions that don't fetch feature flags …') rather than a standing one. The docs mark that behavior as pre-v2.1.268, so the previous present-tense wording overstated the constraint on current versions. Verified against https://code.claude.com/docs/en/cli-reference.

feature-catalog: scope the --system-prompt-snapshot provider caveat to before v2.1.268 (EN + AR)

Applied the same correction to the flag-system-prompt-snapshot entry in src/data/feature-catalog.json and the Arabic counterpart, which seed the global search index and had carried the same present-tense provider caveat. Verified against https://code.claude.com/docs/en/cli-reference.

advanced-features: document the maxEffortLevel effort cap (EN + AR)

Added maxEffortLevel to src/content/modules/advanced-features.md (Planning Mode and Extended Thinking) and the Arabic counterpart. The settings.json key (v2.1.267+) caps the effort level a session can use to low, medium, high, xhigh, or max (max sets no cap); any higher level from /effort, the /model picker, --effort, CLAUDE_CODE_EFFORT_LEVEL, or the model's own default runs at the cap instead. Claude Code applies it before each request, so it holds on every provider including Bedrock, Google's Cloud Agent Platform, and Foundry; when several scopes set a cap the lowest wins, and a per-model cap can go in a model's modelSettings entry. Also added to the feature catalog (EN + AR). Verified against https://code.claude.com/docs/en/settings-reference.

#538

subagents: document the --system-prompt-snapshot flag (EN + AR)

Added --system-prompt-snapshot to src/content/modules/subagents.md and the Arabic counterpart, beside the --append-system-prompt flags. A conversation records its system prompt on the first request and reuses that snapshot on later requests; passing --system-prompt-snapshot off (v2.1.257+) rebuilds it on every request, which matters while you iterate on --append-system-prompt or --append-subagent-system-prompt wording across --continue runs. Recording only applies in sessions that fetch feature flags (claude.ai and Console accounts by default); on Bedrock, Google's Cloud Agent Platform, and Foundry the prompt is rebuilt every request and the flag has no effect. Also added to the feature catalog (EN + AR). Verified against https://code.claude.com/docs/en/cli-reference.

#538

reference-cheatsheet: propagate the /recap, /focus, and /fewer-permission-prompts fact-check fixes (EN + AR)

Propagated the v1.26.44 slash-commands corrections into src/data/reference-cheatsheet.yaml and the Arabic counterpart, which seed the global search index and had continued presenting the stale wording. /recap now separates the on-demand summary from the automatic recap; /focus notes it is only available in fullscreen rendering; /fewer-permission-prompts now says it adds a prioritized allowlist to .claude/settings.json rather than only proposing one. Verified against https://code.claude.com/docs/en/commands.

slash-commands: separate /recap on-demand from the automatic recap (EN + AR)

Rewrote the /recap line in src/content/modules/slash-commands.md (Session tools) and the Arabic counterpart to read 'generates a one-line summary of what happened in the session on demand. A separate automatic recap also appears when you return to the terminal after stepping away.' The previous wording 'Also runs automatically when you return to the terminal after stepping away' conflated the manual /recap command with the separate automatic recap feature. Verified against https://code.claude.com/docs/en/commands.

slash-commands: note that /focus is only available in fullscreen rendering (EN + AR)

Updated the /focus line in src/content/modules/slash-commands.md (Configuration) and the Arabic counterpart to read 'Only available in fullscreen rendering.' The docs explicitly state /focus is only available in fullscreen rendering; the previous wording described only the toggling behavior and would mislead users on the classic renderer into expecting a working command. Verified against https://code.claude.com/docs/en/commands.

slash-commands: correct /fewer-permission-prompts from 'proposes' to 'adds' the allowlist (EN + AR)

Updated the /fewer-permission-prompts line in src/content/modules/slash-commands.md (Configuration) and the Arabic counterpart to read 'adds a prioritized allowlist to .claude/settings.json to reduce future permission prompts.' The previous 'proposes an allowlist' wording implied a manual review step; the docs say the command adds the allowlist itself. Verified against https://code.claude.com/docs/en/commands.

advanced-features: correct Auto Mode supported-model list on Bedrock/Vertex/Foundry (EN + AR)

Rewrote the supported-model sentence in src/content/modules/advanced-features.md (Auto Mode and Permission Control) and the Arabic counterpart to match the permission-modes page. The new wording reads 'only Claude Sonnet 5, Opus 4.7 or later, and the Fable models' instead of 'only Claude Sonnet 5, Opus 4.7, and Opus 4.8 are supported', which had dropped Fable 5 / 5.1 and excluded Opus 5. The 'older models are not supported' sentence was also normalized to use the canonical 'claude-3 models' name. Verified against https://code.claude.com/docs/en/permission-modes.

advanced-features: switch the cloud-session launch example from --remote to --cloud (EN + AR)

Replaced the `claude --remote "your task"` example in src/content/modules/advanced-features.md (Claude Code on the Web and Session Handoff) with `claude --cloud "your task"`, noting that `--remote` is the deprecated alias. The Claude Code on the web page introduces `--cloud` as the canonical form for launching a cloud session from the CLI. Both changes mirrored to Arabic. Verified against https://code.claude.com/docs/en/claude-code-on-the-web.

getting-started: correct PowerShell tool default behavior in Prerequisites (EN + AR)

Rewrote the second sentence of the Windows paragraph in src/content/modules/getting-started.md (Prerequisites) and the Arabic counterpart to match the setup docs: when Git for Windows is installed, the PowerShell tool is available alongside Bash — on by default for claude.ai and Console accounts, and enabled with CLAUDE_CODE_USE_POWERSHELL_TOOL=1 in Amazon Bedrock, Google Cloud's Agent Platform, and Microsoft Foundry sessions. The previous wording described the tool as 'rolling out progressively' and 'not enabled by default for any specific provider', which contradicts the docs. Verified against https://code.claude.com/docs/en/setup.

getting-started: bump npm install Node.js floor from 18+ to 22+ (EN + AR)

Updated the npm install line in src/content/modules/getting-started.md (Installing the CLI) and the Arabic counterpart from '(requires Node.js 18+)' to '(requires Node.js 22+)'. The setup page states 'As of v2.1.198, the npm package requires Node.js 22 or later'; the 18+ floor was the v1.x-era minimum and is well out of date at the current 2.1.263 release. Verified against https://code.claude.com/docs/en/setup.

workflows: correct --from-pr to 'open the session picker filtered to sessions linked to a PR', including quiz Q6 (EN + AR)

Rewrote the --from-pr paragraph in src/content/modules/workflows.md and the full workflows quiz Q6 (question, correct-answer explanation, and the --remote option) in both locales. The docs define --from-pr as opening the session picker filtered to sessions linked to a specific pull request, with sessions linked automatically when Claude creates the pull request — not as bootstrapping a new session by fetching the diff, description, and review comments. Verified against https://code.claude.com/docs/en/cli-reference.

workflows: add custom commands and subagents to the --bare skip list (module + cheatsheet, EN + AR)

Updated the --bare description in src/content/modules/workflows.md and src/data/reference-cheatsheet.yaml to list all eight auto-discovery categories from the docs: hooks, skills, custom commands, subagents, plugins, MCP servers, auto memory, and CLAUDE.md. The previous list omitted custom commands and subagents. Verified against https://code.claude.com/docs/en/cli-reference.

skills: correct paths: scope to 'working with files matching the patterns' (module + terminal, EN + AR)

Updated the paths: frontmatter description in src/content/modules/skills.md and the skills terminal-simulator step: paths: accepts a comma-separated string or YAML list of glob patterns, and Claude loads the skill automatically only when working with files matching the patterns — not when the working directory matches. A glob like src/**/*.ts matches files, not a working directory. Verified against https://code.claude.com/docs/en/skills.

getting-started: spell out Linux/Windows credential storage and the macOS Keychain fallback (EN + AR)

Updated the credential-storage sentence in src/content/modules/getting-started.md: macOS uses the Keychain with a fallback to ~/.claude/.credentials.json (mode 0600) when the Keychain is locked (e.g. over SSH), Linux uses ~/.claude/.credentials.json (mode 0600), and Windows uses %USERPROFILE%\.claude\.credentials.json and inherits the user-profile ACLs. Verified against https://code.claude.com/docs/en/authentication.

reference + feature catalog + terminal: surface the three v2.1.261 features in search (EN + AR)

Added /skill-doctor, --append-subagent-system-prompt/--append-subagent-system-prompt-file, and bashOutputMaxChars to src/data/reference-cheatsheet.yaml (Slash Commands, CLI Flags, and Config Options) and src/data/feature-catalog.json, mirrored to the Arabic files, plus a /skill-doctor command in src/data/terminal-commands.json (and the Arabic terminal) with an entry in /help — so all three are findable through site search and usable in the terminal simulator in both languages.

skills: document /skill-doctor — skill cost and usage report (v2.1.252+) (EN + AR)

Added an 'Auditing Skill Cost with /skill-doctor' section to src/content/modules/skills.md (and the Arabic counterpart) before Skill Visibility Overrides. /skill-doctor reports what each skill costs and how often you invoke it so you can decide which to turn off; it flags skills that have never been invoked and points to where each is turned off, and lists plugins you haven't used recently. It covers session skills other than bundled and enterprise skills. The report opens in the /plugin manager's Stats tab in an interactive session and prints as plain text in -p mode; it needs feature-flag fetching (so it's unavailable in sessions that skip it) and can't run over Remote Control. Requires v2.1.252+. Verified against https://code.claude.com/docs/en/skills.

subagents: document --append-subagent-system-prompt / --append-subagent-system-prompt-file (EN + AR)

Added a paragraph to src/content/modules/subagents.md (and the Arabic counterpart): in non-interactive -p mode you can append shared instructions to every subagent's system prompt, nested subagents included, apart from a forked subagent that reuses the conversation's own prompt. --append-subagent-system-prompt "text" (v2.1.205+) passes the text inline; --append-subagent-system-prompt-file <path> (v2.1.261+) loads the same text from a file when it's too long for the command line. The two flags can't be combined, and both apply only in -p mode. The main agent's equivalents are --append-system-prompt and --append-system-prompt-file. Verified against https://code.claude.com/docs/en/cli-reference.

advanced-features: document bashOutputMaxChars — inline Bash output limit (v2.1.261+) (EN + AR)

Added a 'Sizing Inline Command Output' section to src/content/modules/advanced-features.md (and the Arabic counterpart): Claude Code reads only part of a long Bash result back into the conversation, capped at 30,000 characters by default. Setting bashOutputMaxChars (v2.1.261+) in settings.json raises that up to 128,000 characters — it sizes the inline ceiling and the read-back window together and takes precedence over the older BASH_MAX_OUTPUT_LENGTH environment variable, which Claude Code then ignores. Verified against https://code.claude.com/docs/en/tools-reference and https://code.claude.com/docs/en/settings-reference.

reference + feature catalog: surface the background-session CLI commands in search (EN + AR)

Added claude --bg and a combined claude attach/logs/stop/respawn/rm entry to the CLI Flags section of src/data/reference-cheatsheet.yaml (and the Arabic counterpart), and a cli-background-session entry to src/data/feature-catalog.json (and src/data/ar/feature-catalog.json) mapped to the subagents module, so the background-session commands are findable through site search in both languages.

layout: fix the mobile nav drawer being unscrollable in landscape (issue #528)

The mobile navigation drawer in src/layouts/BaseLayout.astro clamps to max-h-[80vh] but its inner content container used h-full (height:100%), which resolves to auto when the parent has only max-height and no definite height — so the container grew to the full content height and its overflow-y-auto never engaged. On short viewports (mobile landscape) only the top ~80vh was visible and the rest of the menu was unreachable. Changed the inner container from h-full to a definite max-h-[80vh] and added overscroll-contain, so it scrolls within the drawer. Reproduced and verified with headless Chromium at 740×360 (landscape) and 360×740 (portrait): the inner container now reports clientHeight < scrollHeight and scrolls.

subagents: document background sessions — claude --bg and attach/logs/stop/respawn/rm (EN + AR)

Added a paragraph to src/content/modules/subagents.md (and the Arabic counterpart) after the Agent view coverage: claude --bg "prompt" (also --background) starts a session as a background agent and returns immediately, printing the session ID and management commands; combine with --exec to run a shell command as a background job or --agent to background a specific subagent, and it cannot be combined with -p/--print. From the shell, claude attach <id> brings it into the current terminal, claude logs <id> prints recent output, claude stop <id> (also claude kill) stops it, claude respawn <id> restarts a running or stopped session with its conversation intact (--all restarts every running session, e.g. to pick up an updated Claude Code binary), and claude rm <id> drops it from the roster while keeping the transcript so claude --resume can reopen it. This also explains the claude respawn call the module already used in a --json scripting example. Verified against https://code.claude.com/docs/en/cli-reference.

commands: document the Prompt cache (main) line in /usage (v2.1.251+) (EN + AR)

Added a paragraph to the /usage section of src/content/modules/commands.md (and the Arabic counterpart): after the conversation's first API response, /usage adds a Prompt cache (main) line to the Session block summarizing prompt-cache use — request count, share of input tokens served from cache, cache misses (with the time of the last one and tokens re-cached), expected rebuilds from compaction or tool-result clearing, and warm/cold state with the TTL in effect. It covers the main conversation only, not subagents, and status line scripts can read the same numbers from the prompt_cache object. Requires Claude Code v2.1.251 or later. Verified against https://code.claude.com/docs/en/costs (Prompt cache statistics).

commands: document the --permission-prompts CLI flag (EN + AR)

Added a paragraph to src/content/modules/commands.md (and the Arabic counterpart) right after the --eval-dir example, covering the new --permission-prompts flag (Claude Code v2.1.259+): pass none for unattended headless runs where nobody can answer permission prompts, and anything that would prompt is denied automatically while the active permission mode (including auto mode) keeps deciding. Includes a bash example combining --permission-mode auto with --permission-prompts none, and a note that with --output-format stream-json the denied prompts arrive as permission_denied system messages and the final result message lists them under permission_denials. Verified against https://code.claude.com/docs/en/cli-reference.

subagents: document the CLAUDE_CODE_SUBAGENT_MODEL_FORCE env var (EN + AR)

Added a paragraph to src/content/modules/subagents.md (and the Arabic counterpart) right after the --add-dir / --mcp-config paragraph: set CLAUDE_CODE_SUBAGENT_MODEL_FORCE=1 to force one model onto every subagent, teammate, and workflow agent (Claude Code v2.1.257+), ignoring both per-spawn and agent-definition model overrides. The forced model comes from CLAUDE_CODE_SUBAGENT_MODEL or, if that is unset, the main session's model. Includes a bash example pairing CLAUDE_CODE_SUBAGENT_MODEL=sonnet with CLAUDE_CODE_SUBAGENT_MODEL_FORCE=1. Verified against https://code.claude.com/docs/en/env-vars.

project-setup: document the permissions.blockReadsOutsideWorkingDirectories setting (EN + AR)

Added a paragraph to src/content/modules/project-setup.md (and the Arabic counterpart) right after the permissions.additionalDirectories sentence: set permissions.blockReadsOutsideWorkingDirectories: true to keep Claude from reading paths outside the working directories with Read/Grep/Glob/LSP tools in every permission mode including bypassPermissions (Claude Code v2.1.257+); Bash commands that read a matching path through a file command Claude Code recognizes, such as cat, still prompt you even in auto and bypassPermissions mode. Includes a JSON example that turns the option on. Verified against https://code.claude.com/docs/en/settings-reference.

hooks: clarify that `once: true` only works inside skill frontmatter (EN + AR)

The hooks module's description of the `once: true` flag said it runs the hook once per session but omitted the critical caveat that the flag is only honored for hooks declared in skill frontmatter. The module now states that placing the flag in settings.json or in a subagent's frontmatter has no effect, that settings have no session-wide one-shot equivalent, that the flag is per-skill-execution rather than per-session, and names the fail/block/timeout conditions that leave the hook in place for the next match. The matching q5 in the hooks quiz and both bilingual changelog entries were updated alongside to match the corrected module. Mirrored to Arabic. Audit PR against #425 from an earlier cycle, which had tried to demote the levels list instead.

getting-started: document the timeFormat and timeZone interface settings (EN + AR)

Added a paragraph to src/content/modules/getting-started.md (and the Arabic counterpart) right after the preferredNotifChannel sentence, covering timeFormat and timeZone: timeFormat accepts 12-hour, 24-hour, 24-hour-utc, or any strftime pattern (the presets are exposed via /config under Time format, patterns must be set in a settings file), timeZone accepts an IANA name such as "UTC" or "Europe/Dublin" and is ignored when timeFormat is "24-hour-utc", and both keys require Claude Code v2.1.257 or later. Includes a JSON example pairing the 24-hour preset with Europe/Dublin. Verified against https://code.claude.com/docs/en/settings-reference.

commands: document that the fable alias resolves to Fable 5.1 since v2.1.255 (EN + AR)

Added a paragraph to src/content/modules/commands.md (and the Arabic counterpart) covering the default Fable model: the fable alias resolves to Fable 5.1 as of v2.1.255 (Fable 5 before that), ANTHROPIC_DEFAULT_FABLE_MODEL keeps the alias pointing at Fable 5, and on the Anthropic API a claude-fable-5 or claude-fable-5[1m] value saved in user settings is auto-updated to the fable or fable[1m] alias the first time you run v2.1.255+ with the startup line showing (auto-updated) once, while the same value in project, local, or managed settings is left as-is. Verified against https://code.claude.com/docs/en/model-config.

advanced-features: add Fable 5.1 to the thinking-behavior notes alongside Fable 5 (EN + AR)

Updated the MAX_THINKING_TOKENS paragraph in src/content/modules/advanced-features.md (and the Arabic counterpart) so Fable 5.1 sits beside Fable 5 in the two places that list it: the models where thinking cannot be disabled, and the models where adaptive reasoning is always used so positive MAX_THINKING_TOKENS values have no effect. Matches the official statement that Fable 5.1, Fable 5, Sonnet 5, and Opus 4.7+ always use adaptive reasoning. Verified against https://code.claude.com/docs/en/model-config.

feature catalog: add PreModelSwitch and PostModelSwitch hook entries so search surfaces them (EN + AR)

Added hook-pre-model-switch and hook-post-model-switch entries to src/data/feature-catalog.json (and src/data/ar/feature-catalog.json) alongside the other individually-catalogued hook events, so the two new model-transition events are findable through site search in both languages.

hooks: document the PreModelSwitch and PostModelSwitch hook events (EN + AR)

Added a sentence to the Hook Architecture and Configuration section of src/content/modules/hooks.md (and the Arabic counterpart) covering PreModelSwitch (v2.1.251) — fires before Claude Code applies a requested model switch and can block it — and PostModelSwitch (v2.1.251) — fires after the session's model changes, including model restores Claude Code makes on its own when resuming a session. Notes that both receive from_model and to_model fields instead of the model field that only SessionStart gets, and that their matcher is evaluated against the canonical name of the model being switched to (so .*opus.* fires only on switches into an Opus model). Added both events to the module's inline list of events. Verified against https://code.claude.com/docs/en/hooks.

hooks: correct the hook event count from 31 to 33 (EN + AR)

Replaced 'Claude Code supports 31 hook events' with 'Claude Code supports 33 hook events' in src/content/modules/hooks.md and its Arabic counterpart. The current hooks lifecycle table at https://code.claude.com/docs/en/hooks enumerates 33 events — exactly the 31 counted at v2.1.247 plus the new PreModelSwitch and PostModelSwitch. Verified by enumerating the lifecycle table's first column.

changelog: regression tests that a note dropped from the render grouping fails the build

Added a unit test that loads both locales' real changelog data and asserts every entry survives groupEntriesByCategory (catching any future orphaned category, the exact failure that hid the /subtask note), and an explicit assertion that the /subtask note is present. Added the Arabic /changelog route to the RTL smoke suite and a check that the reclassified note renders on both the English and Arabic pages.

changelog: reclassify the orphaned /subtask note from 'terminal' to 'added' so it renders (EN + AR)

The /subtask terminal-simulator note carried category 'terminal', which is not one of the render categories (added/improved/fixed/content/arabic), so groupEntriesByCategory discarded it and it appeared on neither changelog page despite being fully searchable. Reclassified it to 'added' in both locales' changelog data — it documents an addition — without adding a new category value, per the closed-vocabulary decision.

commands: replace the unsupported DISABLE_COMPACT env var with the real /autocompact controls (EN + AR)

DISABLE_COMPACT is not documented in the official env-vars or settings references, so the paragraph claiming DISABLE_COMPACT=1 removes /compact and auto-compaction was removed. Replaced it with the documented mechanism: /autocompact <size> to set the auto-compact window (saved as autoCompactWindow), /autocompact auto to restore the model-tuned window, --autocompact to override for one launch, and CLAUDE_CODE_AUTO_COMPACT_WINDOW for scripts and cloud environments (100K–1M tokens, requires v2.1.221+). Verified against https://code.claude.com/docs/en/model-config.md and cli-reference. Mirrored to Arabic and added to the reference cheatsheet so search catches it.

advanced-features: drop the unsupported CLAUDE_CODE_RESTRICTED env var, keep the --restricted flag (EN + AR)

The --restricted flag is real and documented at https://code.claude.com/docs/en/cli-reference.md (start in restricted mode for an evaluation harness on a shared machine, requires v2.1.248+), but that reference explicitly states there is no CLAUDE_CODE_RESTRICTED environment variable. Removed the '(or CLAUDE_CODE_RESTRICTED=1)' and 'or set the env var' claims from the Restricted Mode section while keeping the accurate flag documentation, and noted the documented eval-harness use case. Mirrored to Arabic.

reference: add the spinnerTipsEnabled and spinnerTipsOverride settings (EN + AR)

Added spinnerTipsEnabled (show the one-line tips under the spinner while Claude works, default true; set false to turn them off) and spinnerTipsOverride (add your own tips to the spinner rotation, or replace the built-in ones — enhanced in v2.1.247 with per-tip entries and a tips file for organization-level tip rotation) to the Config Options section, per code.claude.com/docs/en/settings-reference. Mirrored to Arabic; the site search picks them up automatically.

advanced-features: note that Restricted Mode still loads managed settings and --settings (EN + AR)

Tightened the Restricted Mode section to state that --restricted loads only managed settings and --settings while ignoring the user, project, and local settings files. Verified against code.claude.com/docs/en/cli-reference. Mirrored to Arabic.

workflows: document SELF_HOSTED_RUNNER_CLIENT_LABEL env var (EN + AR)

Added a new 'Self-Hosted Runner' section to src/content/modules/workflows.md (and src/content/modules/ar/workflows.md) describing `claude self-hosted-runner` (introduced in v2.1.224), its polling/execution/reporting model, the new v2.1.248 `SELF_HOSTED_RUNNER_CLIENT_LABEL` env var (and its existing `--client-label` CLI counterpart), and a clear note that the runner is not a data-residency guarantee because the model and control plane remain external. Verified against the v2.1.248 marckrenn CLI surface for the env var, with command attribution reconciled against the v2.1.224 coverage record.

commands: document DISABLE_COMPACT (EN + AR)

Added a paragraph to the Context and Session Management section of src/content/modules/commands.md (and src/content/modules/ar/commands.md) documenting DISABLE_COMPACT=1 as a way to remove /compact and auto-compaction from a session, with a note that the setting is read at session start, a pre-launch example (`DISABLE_COMPACT=1 claude`), and a context-window tradeoff note. Verified against the v2.1.248 marckrenn CLI surface, which lists DISABLE_COMPACT as a new env var at this version.

advanced-features: document --restricted and CLAUDE_CODE_RESTRICTED (EN + AR)

Added a new 'Restricted Mode' section to src/content/modules/advanced-features.md (and src/content/modules/ar/advanced-features.md) at the end of the Auto Mode and Permission Control section. Describes --restricted (and the equivalent CLAUDE_CODE_RESTRICTED=1 env var) as a hardened session that removes the built-in command/code tools and WebFetch unless they're named in --tools, keeps file tools inside the working directory, refuses bypassPermissions, and ignores user, project, and local settings files. Includes two usage examples (one showing the bare flag with -p, and one showing how to expose specific tools via --tools). Verified against the v2.1.248 marckrenn CLI surface, which lists --restricted and CLAUDE_CODE_RESTRICTED as new at this version.

hooks: correct the hook event count from 30 to 31 (EN + AR)

Replaced 'Claude Code supports 30 hook events' with 'Claude Code supports 31 hook events' in the Hook Architecture and Configuration section of src/content/modules/hooks.md. Verified against the v2.1.247 hooks reference lifecycle table (https://code.claude.com/docs/en/hooks.md), which lists 31 events. Mirrored to Arabic by changing 'Claude Code بيدعم 30 hook event' to 'Claude Code بيدعم 31 hook event' in src/content/modules/ar/hooks.md.

slash-commands: correct the /review description (EN + AR)

Replaced the 'fast single-pass, read-only review of a GitHub pull request (requires the `gh` CLI). As of v2.1.223 it's a lighter alias-style entry point to /code-review; reach for /code-review when you want effort levels or fixes' bullet with 'alias of /code-review that reviews the current diff, or a PR number, branch, or path you pass; takes the same effort levels (low–max plus ultra) and flags (--fix, --comment). Before v2.1.223, /review was a separate fast single-pass review of a GitHub pull request.' Verified against the /review row in the v2.1.247 commands reference (https://code.claude.com/docs/en/commands.md), which lists /review as Alias of /code-review with the same effort levels and flags. Mirrored to Arabic.

commands: correct the /effort max and ultracode Q4 explanation (EN + AR)

Replaced the Q4 explanation in src/data/modules/commands/quiz.yaml so the depth list ends at `max` (qualified as model-restricted to Fable 5, Sonnet 5, Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6, session-only, and not accepted in the effortLevel setting), with `/effort ultracode` explained separately as xhigh reasoning plus automatic workflow orchestration (also session-only). Verified against the v2.1.247 model-config and settings-reference docs. Mirrored to Arabic.

advanced-features: correct the MAX_THINKING_TOKENS=0 description (EN + AR)

Reworded the extended-thinking paragraph so MAX_THINKING_TOKENS=0 disables thinking on the Anthropic API on every model except Fable 5 (where thinking also cannot be disabled), and the fixed-budget restriction is moved to the positive-value sentence: on Fable 5, Sonnet 5, and Opus 4.7+ positive values have no effect, and on Opus 4.6 / Sonnet 4.6 you need CLAUDE_CODE_DISABLE_ADAPTIVE_THINKING=1 first to revert to the fixed thinking budget positive values control. Verified against code.claude.com/docs/en/model-config and code.claude.com/docs/en/env-vars. Mirrored to Arabic.

advanced-features: document the /permissions Auto mode tab (EN + AR)

Added a sentence noting that as of v2.1.246, /permissions includes an Auto mode tab for viewing and editing the auto mode classifier rules interactively — an alternative to the claude auto-mode commands and hand-editing the autoMode settings block. Verified against the Claude Code v2.1.246 changelog. Mirrored to Arabic, updated the terminal simulator's /permissions output in both languages, and added a permissions-auto-mode-tab feature-catalog entry to both the English and Arabic catalogs so search surfaces it in each language.

reference: add the modelPicker, promptCacheTtl, and subagentPromptCacheTtl settings (EN + AR)

Added modelPicker (curate the /model picker with an ordered, labeled model list; user or managed, v2.1.243+), promptCacheTtl (main-conversation prompt cache lifetime "5m"/"1h", env-var CLAUDE_CODE_PROMPT_CACHE_TTL, v2.1.242+), and subagentPromptCacheTtl (cache lifetime for subagents and other out-of-conversation requests, env-var CLAUDE_CODE_SUBAGENT_PROMPT_CACHE_TTL, v2.1.242+) to the Config Options section, per code.claude.com/docs/en/settings-reference and code.claude.com/docs/en/prompt-caching. Mirrored to Arabic; the site search picks them up automatically.

reference: add the spellcheck and keybindingFlavor settings (EN + AR)

Added spellcheck (read from user settings or a --settings file, ignored in a project's .claude/settings.json; set { enabled: true } to turn it on) and keybindingFlavor ("readline" makes editing keys follow GNU readline conventions as in Bash — e.g. Ctrl+W deletes back to the previous whitespace; default "classic", requires v2.1.238+) to the Config Options section, per code.claude.com/docs/en/settings-reference. Mirrored to Arabic; the site search picks them up automatically.

reference: restore the Environment Variables and Config Options sections in the Arabic cheatsheet

The Arabic reference cheatsheet was missing the env-vars and config-options categories that exist in English. Translated and added both, so Arabic readers now get the same Environment Variables and Config Options reference as English.

memory: correct autoMemoryDirectory settings-scope claim (EN + AR)

Reworded the autoMemoryDirectory sentence to read the value from any settings scope (user, project, local, policy, or `--settings`), with the project/local workspace trust rule called out instead of a hard rejection. Verified against code.claude.com/docs/en/memory.md. Mirrored to Arabic.

commands: document the Concise built-in output style (EN + AR)

Updated the Output Styles section from four to five built-in styles and added Concise, per code.claude.com/docs/en/output-styles (requires Claude Code v2.1.237+). Mirrored to Arabic.

catalog: list Concise among the built-in output styles (EN + AR)

Added Concise to the Output Styles feature-catalog entry so the search index surfaces it. Mirrored to Arabic.

feedback: Arabic feedback page no longer renders off-screen (EN + AR)

Replaced the feedback form honeypot's left:-9999px positioning with the sr-only clip pattern so it no longer creates ~10000px of horizontal overflow in RTL. The Arabic /ar/feedback page now shows its heading, form, and status columns correctly on mobile.

quiz: RTL-safe off-screen share-card capture (EN + AR)

Wrapped the quiz and self-assessment hidden capture cards in a zero-size overflow:hidden container so their off-screen positioning cannot push the page sideways in RTL, while preserving the cards' full dimensions for image download.

header: mobile controls no longer overflow the viewport (EN + AR)

Moved the language switcher into the mobile drawer, let the logo shrink, and kept the control cluster from shrinking, so the search icon and hamburger menu stay in view and aligned on narrow iPhone screens. The language switcher remains in the top bar on desktop.

commands: ANTHROPIC_DEFAULT_MODEL ignores opusplan/haiku unconditionally (EN + AR)

Removed the unsupported '(the last two only when their respective modes are active)' qualifier from the ignored-tokens sentence. Per code.claude.com/docs/en/model-config#set-a-default-model-for-new-sessions, ANTHROPIC_DEFAULT_MODEL is ignored whenever it is set to default, inherit, opusplan, or haiku.

hooks: correct session_crons source list

Module prose and Q11 quiz now name CronCreate, ScheduleWakeup, and /loop as the actual sources of session_crons, instead of /schedule and plugin monitors. Mirrored to Arabic.

commands + subagents: add the three shipped coverage and audit corrections

Added the two commands references in EN + AR and removed the obsolete --agent-teams activation claim from the Agent Teams guidance in EN + AR.

catalog + cheatsheet: make strictKnownMarketplaces, crossSessionInbound, and dialogExpiry searchable

Added strictKnownMarketplaces to the feature catalog (EN + AR) and corrected the blockedMarketplaces description; added crossSessionInbound, dialogExpiry, and a Cross-session messaging feature entry (EN + AR), plus reference-cheatsheet rows — so the settings the subagents module already explained are now caught by search, which seeds camelCase settings from the catalog and cheatsheet rather than module prose.

plugins: document the owner wildcard for strictKnownMarketplaces and blockedMarketplaces (v2.1.223, close #463)

The plugins module (EN + AR) now describes the owner-wildcard form { source: github, repo: "owner/*" } that allows or blocks every marketplace repo under a GitHub owner in managed settings, with a link to the owner-wildcards reference. Verified against code.claude.com/docs/en/plugin-marketplaces and /settings#owner-wildcards.

subagents: cross-session messaging terminal command and quiz question (v2.1.224, close #464)

Added a /list-agents (alias /peers) command to the terminal simulator (EN + AR) and a subagents quiz question (EN + AR) covering that SendMessage/ListAgents work between independent sessions on macOS/Linux with nothing to enable, and that crossSessionInbound (accept/hold/refuse) controls inbound messages with a dialogExpiry deadline (default five minutes). Verified against code.claude.com/docs/en/cross-session-messaging.

workflows: document /ultrareview --post/--no-post with --no-post as the default (v2.1.227, close #467)

The workflows module (EN + AR) now documents that --no-post is the default (a run posts nothing unless you opt in) and --post posts the finished findings to a github.com PR as a single plain comment from your account, applying only to github.com PR targets. Supersedes auto-opened PR #468, which stated the default backwards. Verified against code.claude.com/docs/en/ultrareview.

a11y: fix all 15 parked jsx-a11y findings and return the rules to errors (#448)

Deleted the eslint.config.js block that downgraded five jsx-a11y rules to warnings for issue #448, after resolving every finding across nine components. The rules (interactive-supports-focus, no-noninteractive-tabindex, no-noninteractive-element-interactions, no-static-element-interactions, click-events-have-key-events) now gate CI as errors again.

a11y: move tablist/radiogroup arrow-key handlers onto their focusable children

TabPanel, ComparisonView, PlaygroundModeToggle (tablists) and QuizBlock (radiogroup) kept the roving-tabindex pattern but moved the arrow/Home/End onKeyDown from the role container onto the focusable tab/radio buttons, where the keydown originates when a tab is focused — the WAI-ARIA APG pattern. No behaviour change for users; the TabPanel Home/End test was updated to fire on a tab.

a11y: accessible pause control for the AutoPlay terminal; role fix for the milestone backdrop

AutoPlayTerminal no longer wraps the whole terminal in a single button role (which would hide the animated output from screen readers). The wrapper is presentational and a dedicated pause/resume button with aria-pressed sits in the title bar as the keyboard/AT control, while click-anywhere-to-pause stays as a mouse convenience. MilestoneModal's click-outside backdrop is now role="presentation"; Escape and the visible Continue button remain the accessible ways to dismiss it.

a11y: keep deliberate scrollable/terminal/diagram affordances behind documented eslint-disables

The horizontally scrollable code regions (ConfigBuilder, TemplateBlock, snippet share page) stay keyboard-focusable per WCAG 2.1.1; InteractiveDiagram's focusable step-through region and its SVG event-delegation container (nodes carry role=button + tabindex) and TerminalSimulator's role="application" are intentional. Each keeps a scoped eslint-disable with a justification comment rather than losing the affordance.

chore: close prompt-injection feedback report #453 with no change

Issue #453, a site feedback submission, contained a prompt-injection attempt instructing the maintainer to recolour the entire site black under a fake 'system update'. Closed as spam/not-planned; no code or content changed.

catalog + cheatsheet: add emojiCompletionEnabled so search catches it

Added an emojiCompletionEnabled entry to the feature catalog (EN + AR) and a Config Options row to the reference cheatsheet (EN) so the setting is discoverable and searchable, matching how cleanupPeriodDays and strictAllowlist are surfaced through the catalog and cheatsheet rather than module prose.

advanced-features: document emoji shortcodes and emojiCompletionEnabled (v2.1.217, close #457)

Added an Emoji Shortcodes section to the Advanced Features module (EN + AR) describing shortcode-to-emoji replacement in the prompt input: `:heart:` becomes ❤️ on the closing colon, `:hea` opens a suggestion popup accepted with Tab or Enter, and the shortcode must start the input or follow a space. On by default; set emojiCompletionEnabled to false in settings.json to disable both the popup and the inline replacement. Verified against code.claude.com/docs/en/interactive-mode.

Add /subtask to the terminal simulator, feature catalog, and reference cheatsheet

Added a /subtask command (EN + AR) to terminal-commands.json /help and a dedicated response, a slash-subtask feature-catalog entry, and a reference-cheatsheet row, so the forked-subagent command the module now references is discoverable and searchable.

subagents: forking is on by default in interactive sessions (v2.1.232), started with /subtask (supersede #445)

Rewrote the forking paragraph in the subagents module (EN + AR) and quiz q5 to match /en/sub-agents: forking is on by default in interactive sessions as of v2.1.232 and off by default in non-interactive -p mode and the Agent SDK; a forked subagent inherits the main session's full conversation context and runs in the background; CLAUDE_CODE_FORK_SUBAGENT=1 also enables it in -p/SDK while =0 disables it everywhere; start one with /subtask (v2.1.212+). Dropped the outdated 'CLAUDE_CODE_FORK_SUBAGENT=1 to enable' and '/fork not a /branch alias' framing. Updated the CLAUDE_CODE_FORK_SUBAGENT feature-catalog entry to describe the override behavior rather than 'to enable'.

skills: distinguish the overall listing budget from the per-entry description cap (reject #443, add skillListingMaxDescChars)

PR #443 wrongly claimed skillListingBudgetFraction and SLASH_COMMAND_TOOL_CHAR_BUDGET aren't skill-listing settings; /en/skills documents both as ways to raise the overall budget (~1% of context), so that content is kept. Added the missing distinction the docs make: the combined description + when_to_use is capped per entry at 1,536 characters, raised with skillListingMaxDescChars — separate from the overall budget. Updated the skills module (EN + AR) and quiz q3 explanation accordingly.

slash-commands: keep the full /effort level list — low/medium/high/xhigh/max/ultracode (reject #444)

PR #444 wanted to reduce /effort to low/medium/high only. /en/model-config lists low, medium, high, xhigh, max for Opus 4.8 and documents ultracode as an additional /effort menu option (session-only). The existing module content, terminal /effort output, and reference cheatsheet were already accurate, so no change; PR closed.

advanced-features + commands: keep the workflow→ultracode rename and the FAST_MODE_OVERRIDE removal at v2.1.160 (reject #436, #437)

Both PRs wanted to change v2.1.160 to v2.1.158 for the dynamic-workflow trigger rename and the CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE removal. The official CHANGELOG lists both lines under the `## 2.1.160` section (directly above the `## 2.1.159` heading), so the existing wording is correct and the audit misread the section boundary. #436 also introduced an internal contradiction — the advanced-features quiz q23 option text stayed v2.1.160 while its explanation was switched to v2.1.158 — which CodeRabbit flagged. Both PRs closed; existing content kept in EN and AR.

getting-started: keep the provider-neutral Windows PowerShell wording (reject #438)

#438 rewrote the Windows PowerShell paragraph to say the PowerShell tool is 'enabled by default' for Bedrock, Vertex, and Foundry. The current docs (/en/setup and the /en/tools-reference PowerShell-tool section) describe availability by Git Bash presence only — automatic when Git Bash is absent, a progressive rollout (opt in/out with CLAUDE_CODE_USE_POWERSHELL_TOOL) when it is present — with no provider-specific default. The module's existing 'not enabled by default for any specific provider' line is docs-accurate; the provider-default claim appears only in the older v2.1.143 CHANGELOG, which the current docs supersede. Rejected; content kept in EN and AR.

skills: keep the skill-listing budget at 1% of the context window (reject #439)

#439 re-proposed the skill-listing budget 1% → 2% change already rejected as #288/#290/#353/#387. The /en/skills docs state verbatim 'The budget scales at 1% of the model's context window'; the 0.02 = 2% figure is given there only as the example for raising the budget via skillListingBudgetFraction, not the default. The module and quiz keep 1% in EN and AR. Rejected.

subagents: keep general-purpose in the built-in list and the v2.1.217–218 depth history (reject #440)

#440 removed `general-purpose` from the built-in-agents list and rewrote the spawn-depth history to 'disabled in v2.1.217–v2.1.218, and one before that'. The /en/sub-agents docs document General-purpose as a built-in agent, so removing it would delete accurate content; and the CHANGELOG shows v2.1.217 disabled nesting (depth 1) and v2.1.219 raised the default to 3 ('was 1'), so the existing 'it was one in v2.1.217–v2.1.218' is correct while the added pre-217 'one before that' claim is unsupported. Rejected; content kept in EN and AR.

plugins: rewrite quiz q1 so the manifest reads as optional, not the 'only required file' (EN + AR)

Quiz q1 asked for the 'only required file' and marked .claude-plugin/plugin.json correct, while a partial audit fix had already changed its explanation to say the manifest is optional — a contradiction where the graded answer rewarded picking a supposedly-required file the explanation called optional. Verified against the official plugins docs (manifest 'optional if components use default locations'), the question now asks what role the manifest plays, the correct answer is that it's optional with auto-discovery from default locations, and all four options and explanations are internally consistent. Also swept the rest of the quiz for the same stale wording: q3's decoy no longer claims plugins 'require the plugin manifest', and q1 option (d) now distinguishes the manifest's identity/configuration fields (name, version, author, userConfig) from the skill and hook definitions that live in separate files, rather than calling it 'only metadata'. Supersedes audit PR #433.

plugins: correct the plugin-manifest template description from 'required' to 'optional' (EN + AR)

The plugin.json template card described the manifest as 'The required .claude-plugin/plugin.json manifest', contradicting the module prose and the official docs. Changed it to 'optional' in both languages so the template surface matches.

subagents: describe agent resumption via SendMessage without an unconfirmed version pin (EN + AR)

The module claimed agents are 'resumable by ID' (vague) and an audit fix pinned the mechanism to 'as of v2.1.76'; that version's official changelog records no Agent resume-parameter removal, and an internal note dated the change to v2.1.77 instead, so the version pin was inconsistent and unconfirmed. Reworded to the observable mechanism — SendMessage({to: agentId}) resumes an agent with its context intact, while a fresh Agent call starts a new conversation — in both languages, dropping the version pin. Supersedes audit PR #434.

subagents: retire the removed per-session cap (CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION) from the module and quiz (EN + AR)

The module prose and quiz q10 decoy still presented the 200-subagent-per-session cap as an active limit. The official CHANGELOG records it was added in v2.1.212 and removed in v2.1.224 (concurrency and depth limits still apply). Reworded the prose so the concurrent cap (default 20) is the remaining companion to the depth limit and the per-session cap reads as removed, and rewrote quiz q10 option (c) so its wrong-answer no longer cites a retired variable as current. Supersedes audit PR #431.

reference surfaces: delete the retired CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION entry from the feature catalog and cheatsheet (EN + AR)

The feature catalog (EN + AR) and the English reference cheatsheet still listed CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION as an active config option with a 200 default, so the catalog, reference page, and site search advertised a variable removed in v2.1.224 — contradicting the corrected module. Removed the entry from all three surfaces.

slash-commands: correct the /review alias version from v2.1.222 to v2.1.223 (EN + AR)

The /review bullet dated its move to a lighter /code-review alias to v2.1.222; the official CHANGELOG records that change in v2.1.223. Fixed the version in both languages and kept the original 'reach for /code-review when you want effort levels or fixes' guidance instead of the auto-fixer's awkward meta-commentary. Supersedes audit PR #430.

reference cheatsheet: correct the plugin manifest entry from required to optional (EN + AR)

v1.24.41 made the plugins module describe .claude-plugin/plugin.json as optional, but the reference cheatsheet still listed it as 'Required plugin manifest' (EN) / 'المطلوب' (AR), contradicting the module. Updated both entries to describe the manifest as optional and to note that components are auto-discovered from default locations when it is omitted.

terminal simulator: add the /effort auto reset line to the Arabic output for EN/AR parity (AR)

The English terminal /effort output ends with 'Use /effort auto to reset to default'; the Arabic output omitted it. Added the Arabic equivalent so both languages document the reset behavior.

slash-commands: add ultracode to /effort and mark max and ultracode as session-only (EN + AR)

The /effort listing in the slash-commands module named low, medium, high, xhigh, and max but omitted ultracode, and marked only max as session-only. The official commands reference states /effort accepts low, medium, high, xhigh, max, or ultracode, that both max and ultracode are session-only, and that auto resets to the model default. Added ultracode, marked both max and ultracode session-only, and marked ultracode session-only in the terminal simulator's /effort output too (EN + AR). Supersedes audit PR #427, whose auto-generated wording redundantly re-listed xhigh.

plugins: correct the manifest from 'the only required file' to optional (EN + AR)

The plugins module described .claude-plugin/plugin.json as 'the only required file.' The official plugins reference states the manifest is optional — if omitted, Claude Code auto-discovers components in their default locations (skills/, agents/, hooks/, .mcp.json, .lsp.json) and derives the plugin name from the directory. Reworded the prose accordingly and changed the '# Required manifest' label in the directory-tree diagram to '# Plugin manifest (optional)'. Supersedes audit PR #426.

hooks: keep the full CLAUDE_EFFORT level list in quiz q9 — rejecting audit PR #425 as based on a stale changelog (EN + AR)

Audit PR #425 proposed reducing the quiz q9 CLAUDE_EFFORT explanation from 'low, medium, high, xhigh, or max' to 'low, medium, high', citing the v2.1.73 CHANGELOG. That source is out of date: the current hooks doc lists the effort.level / $CLAUDE_EFFORT value as 'low, medium, high, xhigh, or max', and the reference cheatsheet documents the same five levels. Removing xhigh and max would have made the quiz wrong. No change made; #425 closed as superseded.

advanced-features: leave the Auto Mode model tiers unchanged — rejecting audit PR #424 as unverified and self-contradictory (EN + AR)

Audit PR #424 rewrote the Auto Mode 'older models never eligible' list but its own finding states the tier list 'should be verified against current docs' and was not. The applied diff also contradicts itself, listing Opus 4.8 among ineligible older models in the same paragraph that names Opus 4.8 as one of the supported models. Per 'if the docs don't support the claim, skip it', the content was left unchanged; #424 closed as superseded.

reference cheatsheet: fix an EN/AR parity bug on /effort max (AR)

The Arabic reference cheatsheet described /effort max as 'لـ Opus بس' (Opus-only) while the English says 'session-only'. Per the official commands reference, max is session-only, not Opus-only. Corrected the Arabic to match.

commands: remove the CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY environment variable — it appears in no official Claude Code source (EN + AR)

The commands module, its templates, a quiz question, the reference cheatsheet, and the feature catalog all described CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY=1 as populating the /model picker from a custom LLM gateway's /v1/models endpoint. The variable is in no official source — not the env-vars reference, the settings page, the CHANGELOG, or the LLM-gateway docs, which describe no gateway model-discovery feature. Removed every reference across both languages (module, template, quiz, cheatsheet, feature catalog). Supersedes audit PR #420, which only softened the module prose with hedge wording.

getting-started: remove the CLAUDE_CODE_DISABLE_ALTERNATE_SCREEN environment variable — it appears in no official Claude Code source (EN + AR)

The getting-started module, a quiz question, the reference cheatsheet, and the feature catalog described CLAUDE_CODE_DISABLE_ALTERNATE_SCREEN=1 as keeping the conversation in the terminal's native scrollback. The variable is in no official source; the only documented alternate-screen variable is CLAUDE_CODE_ALT_SCREEN_FULL_REPAINT (a repaint control, unrelated to disabling the alternate screen). Removed every reference across both languages (module, quiz, cheatsheet, feature catalog). Supersedes audit PR #421, which only softened the module prose with hedge wording.

subagents: keep the 'ultracode sessions are exempt' note and the full 10-question quiz — rejecting audit PR #422 as wrong on both findings (EN + AR)

Audit PR #422 proposed pruning the subagents quiz from 10 to 5 questions ('schema mandates exactly 5') and deleting the 'ultracode sessions are exempt' clause from the CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS description. Both findings are wrong: the content schema imposes no quiz-question count (modules range from 5 to 30 questions), and the official subagents doc states verbatim that sessions with ultracode active are exempt from the concurrent cap. No change made to the subagents content; #422 closed as superseded.

slash-commands: correct the /review description — it's a fast single-pass, read-only review of a GitHub PR (requires the gh CLI), not a 'deeper pass on the change set' (EN + AR)

The module described /review as a 'read-only deeper pass on the change set'. The official commands reference documents it as a distinct bundled skill: a fast single-pass, read-only review of a GitHub pull request that requires the gh CLI. Rewrote the entry to match, noting it's a lighter alias-style entry point to /code-review (v2.1.222) while /code-review remains the choice for effort levels or fixes. Rejects audit PR #416, which had proposed deleting the command entirely — the docs still document /review as its own command, so it was kept and corrected rather than removed. EN + AR modules.

reference cheatsheet: /review reviews a GitHub pull request and needs the gh CLI, not just 'code changes on the current branch' (EN + AR)

The cheatsheet's /review line said 'Review code changes on the current branch with suggestions'. Corrected to match the official commands reference: a fast single-pass, read-only review of a GitHub pull request that needs the gh CLI. EN + AR cheatsheets.

slash-commands quiz: recommend /code-review, not /simplify, for code review (EN + AR)

The /doctor quiz explanation suggested '/review or /simplify' for code review. Per the module itself, /simplify became a separate cleanup-only skill in v2.1.154 that applies fixes without hunting for bugs, so it's no longer the tool for code review. Changed the recommendation to '/review or /code-review' in both quiz files; the correct answer is unaffected.

advanced-features: remove the Ultraplan: Cloud-Powered Planning section — the feature was removed in Claude Code v2.1.222 (EN + AR)

The module documented Ultraplan, which handed a plan to a Claude Code on the web session for cloud drafting. The official anthropics/claude-code CHANGELOG records 'Removed ultraplan feature' in v2.1.222, and it is the only Ultraplan entry in the changelog — a complete removal. Deleted the entire section, including its launch methods, the ◇ ultraplan / ◆ ultraplan ready status badges, the execute-on-web vs. send-to-terminal paths, and the claude.ai/GitHub availability note, from both the EN and AR advanced-features modules. Incorporates audit PR #413 and adds the version bump and changelog it was missing.

project-setup: correct the --add-dir claim — it grants file access only and does NOT load subagent definitions from .claude/agents/ in the added tree (EN + AR)

The module stated that --add-dir loads subagent definitions from .claude/agents/ in the added directory. Per the official /en/commands and /en/permissions docs, most .claude/ configuration is not discovered from an added directory — the flag extends read/edit file access only; CLAUDE.md files and rules still require CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1. Corrected in the EN and AR modules (incorporates audit PR #411).

project-setup quiz: fix the --add-dir explanation, which repeated the same false subagent-loading claim (EN + AR)

The quiz answer explanation for the --add-dir question asserted 'the flag also loads subagent definitions from .claude/agents/ in the added tree'. Rewrote it to match the docs — the flag grants file access only and does not load subagent definitions — in both the EN and AR quiz files, so the quiz no longer contradicts the corrected module (and the terminal simulation and cheatsheet, which already stated .claude/ config is not loaded).

getting-started: name Debian 10+ and Alpine Linux 3.19+ as supported distros; keep Ubuntu 20.04+ (EN + AR)

The prerequisites line said 'Ubuntu 20.04+ (and other modern Linux distros)'. Per the /en/setup System requirements section, added the specific distros the docs name — Debian 10+ and Alpine Linux 3.19+ — while keeping Ubuntu 20.04+ (the docs list 20.04, so the audit's proposed 22.04 change was rejected). This matches the getting-started install tabs, which already list Ubuntu 20.04+, Debian 10+, and Alpine 3.19+. Applied to EN and AR (incorporates the supported part of audit PR #410).

subagents: cover nested subagents — default depth three (v2.1.219, was one), the Agent-tool withholding at the depth limit, and CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH (EN + AR)

The Subagents module never mentioned that a subagent can spawn subagents of its own. Per the official /en/sub-agents 'Let subagents spawn their own subagents' section, added that nesting is allowed up to three layers below the main conversation by default (raised to three in v2.1.219, having been one in v2.1.217–v2.1.218), that at the depth limit the Agent tool is withheld from every subagent except a fork, and that CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH (v2.1.217+) sets the number of layers or 1 disables nesting. Applied to the EN and AR modules.

subagents: cover the concurrent (20) and per-session (200) subagent limits and their env vars (EN + AR)

Alongside nesting depth, documented the two other limits the docs describe: CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS (default 20 running at once, v2.1.217+, ultracode sessions exempt) and CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION (default 200 total per session, v2.1.212+, reset by /clear). Added to the module prose (EN + AR), the feature catalog (EN + AR) and the reference cheatsheet (EN) so all three are discoverable through search.

subagents quiz: add q10 on the default nesting depth and how to change it (EN + AR)

Added a tenth question to the Subagents quiz testing that a subagent nests up to three layers below the main conversation by default (v2.1.219) and that CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH changes the number of layers (1 to disable), with decoys that conflate depth with the separate concurrent (20) and per-session (200) limits. Added to both the EN and AR quiz files.

Arabic parity for the nested-subagents coverage

Every addition was applied to the Arabic side alongside the English: the nested-subagents-and-limits paragraph in ar/subagents.md, the q10 quiz question in ar/quiz.yaml, and the three env-var entries in ar/feature-catalog.json. The reference cheatsheet is English-only (the Arabic cheatsheet has no env-vars section), so the cheatsheet entries are EN-only, consistent with prior passes.

commands: Fast Mode supports Opus 5 and Opus 4.8, not Opus 4.7 — fixed the self-contradiction the auto-fix left, plus the default model and rate-limit pool (EN + AR)

Audit PR #400 corrected the deprecation-date wording but left the Fast Mode paragraph contradicting itself: it listed Opus 4.7 as a supported model in the same sentence that then said it had been removed in v2.1.219. The official CHANGELOG (v2.1.219) states 'Removed Opus 4.7 from fast mode; /fast now applies to Opus 5 and Opus 4.8' and that Opus 5 is 'now the default Opus model'. Rewrote the supported-models clause to Opus 5 (default Opus model since v2.1.219) and Opus 4.8, dropping Opus 4.7; changed the auto-switch default from Opus 4.8 to Opus 5; and updated the fast-mode rate-limit pool from 'shared across Opus 4.8 and 4.7' to 'shared across Opus 5 and Opus 4.8'. Incorporates and closes audit PR #400 and the remaining Codex line-70 finding it did not cover.

project-setup quiz: --add-dir loads .claude/agents/ subagent definitions; CLAUDE.md and rules still need CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1 (EN + AR)

The q7 option-(a) explanation claimed --add-dir 'only grants file access; .claude/ configuration inside the added paths is not loaded', which directly contradicts the project-setup module body — it already states --add-dir loads subagent definitions from .claude/agents/ in the added tree, while CLAUDE.md files and rules require CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1 to take effect (per the /en/sub-agents docs). Corrected the quiz explanation to match. Incorporates and closes audit PR #401.

Arabic parity for both fixes

Both corrections were applied to the Arabic side alongside the English: the Fast Mode supported-models, default-model and rate-limit-pool wording in ar/commands.md, and the project-setup q7 quiz explanation in ar/quiz.yaml.

Windows PowerShell tool is an opt-in progressive rollout, not default-on for Bedrock/Vertex/Foundry — getting-started, advanced-features, terminal, catalog, cheatsheet (EN + AR)

Our Windows setup guidance said the PowerShell tool 'is enabled by default on Windows for Bedrock, Vertex, and Foundry users — set CLAUDE_CODE_USE_POWERSHELL_TOOL=0 to opt out.' The current official /en/setup docs no longer describe any default-on provider: 'When Git for Windows is installed, the PowerShell tool is rolling out progressively as an additional option alongside Bash. Set CLAUDE_CODE_USE_POWERSHELL_TOOL=1 to opt in or 0 to opt out.' Rewrote the getting-started prose to lead with the opt-in rollout (=1 to opt in, =0 to opt out, not enabled by default for any specific provider), and applied the same correction to the advanced-features module, the advanced-features terminal-steps description, the feature catalog entry, and the reference cheatsheet entry. Incorporates audit PR #398 and extends it to the four other places the stale default-on claim survived.

Arabic parity for the PowerShell rollout correction

Every change was applied to the Arabic side alongside the English: the opt-in progressive-rollout wording in the getting-started and advanced-features modules and the advanced-features terminal-steps description. The feature catalog and reference cheatsheet PowerShell entries are English-only content (the Arabic cheatsheet carries no CLAUDE_CODE_USE_POWERSHELL_TOOL entry), so no further Arabic parity fixes were needed there.

advanced-features: cover parentSettingsBehavior, sandbox.network.strictAllowlist and the DirectoryAdded hook — EN + AR

The module (and the parentSettingsBehavior quiz question) referenced settings the prose did not explain. Added three documented items: parentSettingsBehavior (v2.1.133, admin-tier), with 'merge' clarified as restrictive layering — parent-supplied settings sit beneath admin policy through a restrictive-only filter so they can tighten but not loosen it, and the setting has no effect without an admin tier; sandbox.network.strictAllowlist (v2.1.219), written with its full setting path, which denies all hosts not in allowedDomains without prompting; and a Hooks section for DirectoryAdded (v2.1.219), described as firing when /add-dir or the SDK register_repo_root registers an additional working directory mid-session — a directory-registration event, not a change of the current directory (distinct from the CwdChanged hook). Incorporates audit PR #394 with the CodeRabbit and Codex review refinements.

commands: Ctrl+O toggles the transcript viewer, not 'verbose mode' — EN + AR, site-wide

The official interactive-mode docs define Ctrl+O as 'Toggle transcript viewer', which shows detailed tool usage and execution — not a 'verbose mode'. Corrected the commands module description and, for consistency, every place the Ctrl+O = 'verbose mode / same as /debug' conflation appeared: the reference cheatsheet Ctrl+O and /debug entries, the /debug terminal simulation, and the commands, slash-commands and subagents quiz explanations. /debug is corrected to 'enable debug logging' per the /en/commands docs. The auto-fix's added claim that Ctrl+Shift+V or --include-partial-messages enables verbose mode was rejected: Ctrl+Shift+V is undocumented and --include-partial-messages is a streaming-JSON flag (requires --print and --output-format stream-json), not an interactive verbose toggle. Incorporates the verified part of audit PR #395.

feature catalog: add sandbox.network.strictAllowlist and the DirectoryAdded hook for search

Added catalog entries for sandbox.network.strictAllowlist (v2.1.219) and the DirectoryAdded hook (v2.1.219) so the two newly documented features are discoverable through the site search alongside their module coverage.

workflows quiz: keep the v2.1.120 attribution for the claude ultrareview CLI subcommand (rejected audit PR #396)

Audit PR #396 tried to change the workflows q7 explanation from v2.1.120 to v2.1.111, but that explanation is about the `claude ultrareview [target]` CLI subcommand running non-interactively in CI. The v2.1.121 CHANGELOG documents that subcommand under v2.1.120 ('Added `claude ultrareview [target]` subcommand to run /ultrareview non-interactively from CI or scripts'); v2.1.111 introduced the interactive /ultrareview slash command, a different feature. v2.1.120 is a real documented version, so the original attribution is kept and #396 is closed without applying its change.

Arabic parity for every change in this pass

Every change was applied to the Arabic side alongside the English: the parentSettingsBehavior / strictAllowlist / DirectoryAdded additions in the advanced-features module, the Ctrl+O transcript-viewer and /debug debug-logging corrections across the commands/slash-commands/subagents quizzes, the reference cheatsheet, and the terminal simulation.

commands: /code-review runs as a background subagent as of Claude Code v2.1.218 — EN + AR

The commands module and the /code-review quiz explanation described the command without noting that, as of v2.1.218, it runs as a background subagent so review work no longer fills the conversation. Per the Claude Code CHANGELOG v2.1.218 ('Changed /code-review to run as a background subagent, so review work no longer fills your conversation'), added that behaviour to the commands module prose and the quiz explanation. Applied to the EN + AR module and quiz. Incorporates audit PR #392.

advanced-features + workflows quizzes: replace the non-existent --no-conversation flag with --no-session-persistence — EN + AR

The advanced-features headless-mode Q4 decoy and the workflows CI/CD Q1 decoy both named --no-conversation, a flag that does not exist. The official CLI reference lists --no-session-persistence as the print-mode (claude -p) flag that disables session saving and has no --no-conversation entry. Corrected both the decoy option text and its explanation in each quiz so the decoy teaches the real flag while remaining an incorrect answer (the CI/CD flag is still --permission-mode bypassPermissions). Incorporates audit PR #391 and resolves the CodeRabbit + Codex review notes that its explanation text was left stale.

Arabic parity for every change in this pass

Every change was applied to the Arabic side alongside the English: the v2.1.218 /code-review background-subagent note (commands module + quiz) and the --no-conversation → --no-session-persistence correction (advanced-features and workflows quizzes, option text and explanation).

PowerShell tool is enabled by default on Windows for Bedrock/Vertex/Foundry — getting-started, advanced-features, terminal (EN + AR)

The Windows setup guidance said the PowerShell tool 'is rolling out progressively' and led with CLAUDE_CODE_USE_POWERSHELL_TOOL=1 to opt in. Per the Claude Code CHANGELOG, the tool is now enabled by default on Windows for Bedrock, Vertex, and Foundry users — opt out with CLAUDE_CODE_USE_POWERSHELL_TOOL=0 — and only rolls out progressively for other providers. Rewrote the wording to lead with the default-on behavior in the getting-started module and, for consistency across the site, in the advanced-features module and the advanced-features terminal-steps description as well. Incorporates audit PR #386 and a Codex review note about the remaining stale references.

workflows: /ultrareview was added in Claude Code v2.1.111 — EN + AR

The workflows module attributed /ultrareview to 'new in v2.1.86, highlighted in v2.1.112'. The CHANGELOG's introducing entry ('Added /ultrareview for running comprehensive code review in the cloud…') is under v2.1.111, with no matching v2.1.86 or v2.1.112 addition. Corrected the attribution to 'added in v2.1.111', aligning the workflows module with the skills module's existing v2.1.111 reference. Incorporates the first finding of audit PR #389.

changelog: correct the stale v1.22.4 /ultrareview attribution to v2.1.111

The historical v1.22.4 changelog entry still attributed /ultrareview to v2.1.86, which contradicted this release's correction to v2.1.111. Updated that entry's version references (EN + AR) so the public changelog is internally consistent on the official v2.1.111 attribution. Addresses a CodeRabbit data-integrity review comment.

Rejected audit PRs #387 (skills), #388 (slash-commands), and #389's second finding

#387 re-proposed the skill-listing budget 1% → 2% change (already rejected as #288/#290) — the /en/skills docs say 'The budget scales at 1% of the model's context window' — and tried to trim skill `effort` to low/medium/high, but /en/model-config and the skills frontmatter reference both list `low`, `medium`, `high`, `xhigh`, `max`; rejected, original kept. #388 tried to drop `max` from the /effort values, but /en/model-config confirms `max` is a valid (deepest, session-only) effort level; rejected. #389's second finding claimed `disableWorkflows` / `CLAUDE_CODE_DISABLE_WORKFLOWS=1` don't exist, but the workflows docs 'Turn workflows off' section documents both verbatim; rejected, and the original module wording plus the reference cheatsheet entry were kept. All of #386–#389 are closed in favour of this consolidated PR.

Arabic parity for every change in this pass

Every change in this pass was applied to the Arabic side alongside the English: the PowerShell default-on wording (getting-started module, advanced-features module, and advanced-features terminal-steps), the /ultrareview v2.1.111 attribution (workflows module), and the corrected v1.22.4 changelog entry. No Arabic quiz, template, cheatsheet, or catalog content asserted the corrected facts, so no further Arabic parity fixes were needed.

project-setup: note that plugin `monitors` support was added in Claude Code v2.1.105 — EN + AR

The guidance on experimental plugin-manifest fields (`monitors` and `themes` under `experimental: {}`) now dates the `monitors` field to its introducing release. Per the Claude Code CHANGELOG, background monitor support for plugins landed in v2.1.105; the `experimental` nesting requirement followed in v2.1.129. Applied to the EN + AR module. Incorporates audit PR #384.

advanced-features: Advisor Tool is enabled via /advisor, advisorModel, or --advisor — not a fabricated env var — EN + AR

The module and both quiz files stated the Advisor Tool is enabled via CLAUDE_CODE_ENABLE_EXPERIMENTAL_ADVISOR_TOOL. Per the official advisor docs (code.claude.com/docs/en/advisor), no such env var exists — the advisor is enabled by setting a model through the /advisor command, the advisorModel setting, or the --advisor flag, and CLAUDE_CODE_DISABLE_ADVISOR_TOOL is the only advisor env var (it disables the tool). Corrected the claim across the EN + AR module and EN + AR quizzes. Incorporates audit PR #382.

advanced-features (AR): correct 'disable' wording and a typo in the advisor content

The Arabic module and quiz described CLAUDE_CODE_DISABLE_ADVISOR_TOOL as 'بينعّم الأداة' (softens the tool) rather than 'بيعطّل الأداة' (disables the tool), and the quiz carried a 'بيتتفعّل' → 'بتتفعّل' typo. Both fixed per review feedback on PR #382.

plugins: document LSP restartOnCrash/shutdownTimeout and settings.json supported keys — EN + AR

Per the plugins reference (code.claude.com/docs/en/plugins-reference), the LSP `restartOnCrash` and `shutdownTimeout` fields require v2.1.205+ (before which setting either skipped that LSP server at startup), and a plugin's `settings.json` supports only the `agent` and `subagentStatusLine` keys — all other fields are silently ignored. Added both to the plugins module. The unverified `--plugin-dir (v2.1.145+)` tag from audit PR #375 was NOT added (no doc support). Partially incorporates #375.

memory: expand /init multi-phase flow, add decline-import note, clarify no-paths rules — EN + AR

Per the memory docs (code.claude.com/docs/en/memory): expanded the `CLAUDE_CODE_NEW_INIT=1` description (asks which artifacts to set up, explores with a subagent, presents a reviewable proposal); noted that declining a first-time external-import dialog leaves imports disabled and the dialog doesn't reappear; added that files are concatenated in load order (managed → user → project → local, closest read last); and clarified in the quiz that rules without a `paths` field load unconditionally and apply to all files. Partially incorporates #374 — its 'whichever is smaller' change was dropped (docs say 'whichever comes first') and its '~v2.1.198' tag was dropped (misattributed to symlink matching).

subagents: built-in agents CAN be name-overridden by a custom subagent — EN + AR

The module claimed built-in subagents are 'always available and not part of the name-override priority system'. The sub-agents docs (code.claude.com/docs/en/sub-agents) say the opposite: 'A user or project subagent named Explore overrides the built-in and keeps its own model field.' Rewrote the sentence to describe the override behavior. Incorporates audit PR #378.

project-setup: --add-dir loads subagent definitions from .claude/agents/ — EN + AR

The `--add-dir` paragraph said the flag 'only grants file access, not the rest of .claude/ configuration'. Per the docs, `--add-dir` also scans a `.claude/agents/` folder in the added directory and loads those subagents; CLAUDE.md files and rules still require `CLAUDE_CODE_ADDITIONAL_DIRECTORIES_CLAUDE_MD=1`. Corrected the sentence. Incorporates audit PR #376.

workflows: routines require Claude Code on the web enabled — EN + AR

The routines availability line named the plans (Pro/Max/Team/Enterprise) but omitted the prerequisite. The routines docs (code.claude.com/docs/en/routines) state routines are available on those plans 'with Claude Code on the web enabled'. Added the clause. Incorporates audit PR #379.

Rejected audit PRs #372 (ultracode/workflow rename) and #373 (MCP scope reduction); closed no-op #377

#372 flipped the ultracode↔workflow rename direction, contradicting the current dynamic-workflow keyword and leaving the paragraph self-contradictory — rejected, original kept. #373 tried to reduce the three MCP scopes to two and claim local scope is stored in `.mcp.json` and private, but the MCP docs (code.claude.com/docs/en/mcp) list three scopes with Local in `~/.claude.json` and Project in the shared `.mcp.json`; only its `which channel plugins` allowedChannelPlugins wording fix was kept. #377 (skills) had an empty diff and was closed as a no-op. All of #372–#379 are closed in favor of this consolidated PR.

Arabic parity for every change in this pass

Mirrored all module edits (subagents, project-setup, workflows, plugins, memory, mcp) and the memory quiz clarification in the Arabic content and quiz files, matching the existing Egyptian-Arabic style.

slash-commands quiz: correct the /branch Q2 explanation's stale /fork framing

Q2's explanation called `/branch` '(formerly /fork)', implying `/branch` was renamed from `/fork`. Per the commands and sub-agents docs, `/fork` was an alias for `/branch` only before v2.1.161; from v2.1.161 `/fork` is a distinct command that hands a side task to a background forked subagent. Reworded the explanation to describe `/branch` on its own terms and note that `/fork` is now a separate command, not an alias — consistent with the subagents module (src/content/modules/subagents.md).

Rejected audit PRs #367 (plugins prune/autoremove) and #368 (slash-commands /fork alias)

Both auto-generated audit PRs contradicted the official docs and were closed. #367 would have removed the correct 'aliased autoremove' note on `claude plugin prune` — the Plugins reference lists 'Aliases: autoremove' for that command. #368 would have added a false '`/fork` is its alias' claim to the `/branch` bullet — `/fork` is a distinct background-subagent command from v2.1.161. The repo's existing module content on both topics is already accurate and left unchanged.

Arabic parity confirmed for the /branch quiz explanation

The Arabic slash-commands quiz Q2 already omits the '/fork' parenthetical, so the EN fix brings the two languages back into agreement; no Arabic change was needed for this correction.

advanced-features: document the screen-reader accessibility mode — EN + AR

Per the settings docs (code.claude.com/docs/en/settings), Claude Code exposes a screen-reader mode as the `axScreenReader` setting (v2.1.181+), the `CLAUDE_AX_SCREEN_READER=1` environment variable, and the `--ax-screen-reader` flag — the variable and flag take precedence over the setting. It renders flat text with no decorative borders or animations for assistive technology, and always uses the classic renderer so the fullscreen/tui preference has no effect while active. Added the environment variable to the advanced-features env-vars section, next to the existing accessibility-related CLAUDE_CODE_NATIVE_CURSOR entry.

reference + catalog: add the --ax-screen-reader flag and CLAUDE_AX_SCREEN_READER env var

Added the `--ax-screen-reader` flag to the CLI-flags cheatsheet (EN + AR), the `CLAUDE_AX_SCREEN_READER` variable to the EN env-vars cheatsheet, and an `env-ax-screen-reader` feature-catalog entry (EN + AR) so the accessibility feature is discoverable through catalog search in both languages.

Merged audit PRs #364 (mcp Channels wording) and #365 (plugins storage wording)

Both auto-generated audit PRs were verified against the official docs, CodeRabbit-clean, and Cloudflare-green. #364 clarifies that `--channels` was added in v2.1.80 as a research preview and suggests checking `claude --version`; #365 clarifies `${CLAUDE_PLUGIN_DATA}` is a shell-expansion variable for persistent state and reframes `claude plugin tag` as 'introduced in v2.1.118'. Squash-merged with clearer commit titles (#364's original title was flagged as too vague by CodeRabbit).

Arabic parity for the screen-reader-mode additions

Mirrored the advanced-features env-var paragraph, the CLI-flags cheatsheet entry, and the feature-catalog entry in Arabic.

mcp: document the WebSocket (ws) transport — EN + AR

Per the MCP docs (code.claude.com/docs/en/mcp), WebSocket is a fourth MCP transport for remote servers that push events unprompted over a persistent bidirectional connection. Added `ws` to the transport-selection guidance, noting it must be configured with `type: "ws"` in `.mcp.json` (not `claude mcp add --transport`, which doesn't accept `ws`) and supports neither OAuth nor `--transport`. Consolidates audit PR #360.

mcp: correct the WebSocket feature-catalog config — EN + AR

The WebSocket MCP Transport feature-catalog entry described the config as `transport: websocket`, which is not how it's configured. Corrected to `type: "ws"` in `.mcp.json`, matching the module and the official docs.

advanced-features: Auto Mode is available by default on all four providers — EN + AR

Per the permission-modes docs (code.claude.com/docs/en/permission-modes), Auto Mode is now available by default on the Anthropic API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Foundry; the `CLAUDE_CODE_ENABLE_AUTO_MODE=1` opt-in that applied on the non-Anthropic providers in v2.1.158–v2.1.206 was removed in v2.1.207 (still accepted for compatibility, no effect). Also corrected the supported-model list on the three non-Anthropic providers to Claude Sonnet 5, Opus 4.7, and Opus 4.8 — the previous text and PR #359 both dropped Sonnet 5. Consolidates audit PR #359.

Rejected audit PR #361's --max-turns → --runs replacement

#361 proposed replacing `--max-turns` with a `--runs <n>` flag across the workflows module and templates (EN + AR), citing a third-party changelog. The official CLI reference (code.claude.com/docs/en/cli-reference) confirms `--max-turns` exists and lists no `--runs` flag, so the change would have introduced a non-existent flag. The existing `--max-turns` content is correct and left unchanged; #361 is closed.

Arabic parity for the mcp and advanced-features updates

The Arabic mcp and advanced-features modules and the Arabic feature-catalog mirror all three EN changes — the WebSocket (ws) transport guidance, the corrected WebSocket config, and the Auto Mode default-availability and supported-model updates.

mcp: name the WaitForMcpServers fallback when tool search is disabled

Per the MCP docs (code.claude.com/docs/en/mcp), tool search is on by default, and in configurations without it — Google Cloud's Agent Platform, a custom ANTHROPIC_BASE_URL, or ENABLE_TOOL_SEARCH=false — Claude uses the WaitForMcpServers tool instead. Replaced the older 'off by default on Vertex AI / non-first-party proxy' aside with this doc-verified phrasing. Consolidates audit PR #355.

memory: @import loads files into context at launch and still consumes tokens

Per the memory docs (code.claude.com/docs/en/memory), the @path/to/file import syntax expands imported files and loads them into context at launch alongside the CLAUDE.md that references them — they still consume context tokens. Replaced the misleading 'reference rather than duplicate' phrasing. Also clarified that ./CLAUDE.md and ./.claude/CLAUDE.md are two alternative supported locations for the project scope, not one shared file. Consolidates the doc-supported part of audit PR #356.

plugins: correct marketplace sources (npm is not one) — EN + AR

Per the discover-plugins docs (code.claude.com/docs/en/discover-plugins), plugin marketplaces source from GitHub repositories (owner/repo), git URLs, local paths, and remote marketplace.json URLs — not npm. Removed the incorrect npm claim in the EN module and added the corrected sentence to the AR module, which previously lacked it. Also reordered the Arabic plugins quiz (q5 cache cleanup, q6 version tag, q7 subagent restrictions) to match the English sequence. Consolidates audit PR #357.

Rejected audit PR #356's 'three memory locations' reframing

#356 proposed collapsing the CLAUDE.md scopes to 'three truly distinct locations: managed policy, user/project instructions, and local instructions.' The memory docs list four distinct scopes and keep user (~/.claude/CLAUDE.md, personal) separate from project (./CLAUDE.md, team-shared), so this reframing is not supported and was dropped; only the accurate clarification that ./CLAUDE.md and ./.claude/CLAUDE.md are two alternative supported locations for the project scope (not one shared file) was kept.

Arabic parity for the mcp, memory, and plugins corrections

The Arabic mcp and memory modules mirror the WaitForMcpServers fallback and the @import context-cost correction; the Arabic plugins module gains the corrected marketplace-sources sentence; and the Arabic plugins quiz is realigned to the English question order (q5–q7).

commands: fast mode supports Opus 4.8 and 4.7 only — Opus 4.6 fast mode is retired

Per the fast-mode docs (code.claude.com/docs/en/fast-mode), fast mode is supported on Opus 4.8 and Opus 4.7 only; earlier Opus models like 4.6 are no longer supported. Replaced the stale 'supports Opus 4.6, removed ~30 days after the Opus 4.8 launch' copy with the doc-supported facts (Opus 4.7 fast mode deprecated June 25, 2026, removed July 24, 2026, no fallback), and corrected the rate-limit pool to 'shared across Opus 4.8 and 4.7'. Avoided the unverified v2.1.176 version claim from audit #352.

commands + terminal + cheatsheet: Shift+Tab cycles default → acceptEdits → plan

Per the permission-modes docs (code.claude.com/docs/en/permission-modes), Shift+Tab cycles `default` → `acceptEdits` → `plan`; `auto`, `bypassPermissions` (added by `--permission-mode bypassPermissions`, `--dangerously-skip-permissions`, or `--allow-dangerously-skip-permissions`), and `dontAsk` are not in that default cycle. Rewrote the commands module prose to include the `--allow-dangerously-skip-permissions` variant (missed by audit #352) and fixed the wrong order (`default → plan → acceptEdits → auto`) in the terminal `/permissions` output, the project-setup terminal step, and the reference cheatsheet.

Rejected audit PR #351 (enforceAvailableModels) and #353 (/fewer-permission-prompts rename, skills budget 1%→2%)

#351's 'managed-only, cannot be set in user or project settings' qualifier for `enforceAvailableModels` is not supported by the settings docs, which describe a setting settable in any scope; the existing prose stays. #353 re-proposed two already-rejected regressions — the `/less-permission-prompts` rename (live skill and commands reference use `/fewer-permission-prompts`) and the skills budget 1%→2% change (docs say the budget scales at 1% of the context window) — so both are declined.

Arabic parity for the fast-mode and Shift+Tab corrections

The Arabic commands module, project-setup terminal step, terminal `/permissions` output, and reference cheatsheet mirror the fast-mode Opus 4.8/4.7 update and the corrected `default → acceptEdits → plan` cycle order.

mcp: alwaysLoad: true also blocks startup until that server connects (5-second connect cap)

Per the MCP docs (code.claude.com/docs/en/mcp), `alwaysLoad: true` doesn't only exempt a server from tool-search deferral — it also blocks startup until that server connects, capped at the standard 5-second connect timeout, because the tools must exist when the first prompt is built. Other configured servers keep connecting in the background. Added this to the MCP module, which previously described only the deferral-exemption behavior.

Arabic parity for the alwaysLoad startup-blocking note

The Arabic MCP module mirrors the added note (alwaysLoad blocks startup on that server's connect, 5-second cap, other servers keep connecting in the background).

hooks + subagents: 'auto' is not an effort level (hook input and subagent frontmatter list low/medium/high/xhigh/max)

Per the hooks docs (code.claude.com/docs/en/hooks) and sub-agents docs (code.claude.com/docs/en/sub-agents), the hook-input `effort` object and the subagent frontmatter `effort` field each accept only `low`, `medium`, `high`, `xhigh`, `max`. Removed the invalid `auto` value from both modules. `auto` remains a valid `/effort` command argument (reset to model default) in the commands and slash-commands modules, so those are unchanged.

Arabic parity for the effort-level correction

The Arabic hooks and subagents modules mirror the corrected effort-level list (low/medium/high/xhigh/max, no auto).

workflows + cheatsheet: --bare is a startup-speed flag, not an output-formatting one

Per the headless docs (code.claude.com/docs/en/headless), `--bare` reduces startup time by skipping auto-discovery of hooks, skills, plugins, MCP servers, auto memory, and CLAUDE.md, so only explicitly-passed flags and settings take effect. Replaced the misleading 'cleanest scripted output' framing in the workflows module and the 'Cleanest output... no formatting or decoration' description in the reference cheatsheet.

Arabic parity for the --bare correction

The Arabic workflows module and reference cheatsheet mirror the corrected --bare description (startup speed and reproducibility over output cleanliness).

Ctrl+G opens the proposed plan in your editor when approving in plan mode

Per the permission-modes docs (code.claude.com/docs/en/permission-modes), pressing `Ctrl+G` at the plan-approval prompt opens the proposed plan in your default text editor to edit before Claude proceeds. Kept this correct content (audit PR #334 had wrongly removed it) and refined two surfaces for precision: the reference cheatsheet now separates the general behavior (opens your prompt) from the plan-mode behavior, and the terminal simulator's /plan output no longer suggests Ctrl+G exits plan mode.

Arabic parity for the Ctrl+G plan-edit clarification

The Arabic advanced-features module, reference cheatsheet, terminal simulator /plan output, and guided terminal step mirror the clarified Ctrl+G plan-edit content.

getting-started: the Desktop App now supports Linux in beta (Ubuntu 22.04+/Debian 12+)

Per the official docs (code.claude.com/docs/en/desktop-linux), the Claude desktop app is in beta on Linux — Ubuntu 22.04+ or Debian 12+, x86_64/arm64, installed via Anthropic's apt repository or a downloaded .deb — with Computer Use, dictation, and Fedora/RHEL not yet included. Replaced the outdated 'not available on Linux' line in the getting-started module, using the full docs URL rather than a root-relative link that would 404 on the static site.

Arabic parity for the Desktop App Linux-beta correction

The Arabic getting-started module mirrors the Desktop-App-on-Linux-beta correction, including the beta's known gaps and the full docs URL.

subagents: /agents no longer opens an interactive panel (v2.1.198)

Per the official docs (code.claude.com/docs/en/agents), as of v2.1.198 /agents no longer opens a panel — it prints a notice pointing to the subagent file locations, and custom subagents are created or edited by asking Claude or editing the files directly. Corrected across the subagents module prose, the reference cheatsheet, the terminal simulator's /agents output, the subagents guided terminal step, three subagents quiz explanations, and the /agents feature-catalog entry.

Arabic parity for the /agents behavior correction

The Arabic subagents module, reference cheatsheet, terminal simulator, guided terminal step, subagents quiz, and feature catalog mirror the /agents-prints-a-notice correction.

subagents: the built-in Explore agent inherits the main session model (capped at opus), not Haiku

Per the Claude Code CHANGELOG (v2.1.198), the built-in Explore agent now inherits the main session's model, capped at opus, instead of running on Haiku. Corrected across the subagents module prose, all four subagents quiz options (the correct answer and the a/c/d distractor explanations), the skills quiz explanation for `agent: Explore`, the reference cheatsheet, the terminal simulator's `claude agents` output, and the Explore feature-catalog entry.

slash-commands: kept /fewer-permission-prompts (rejected audit PR #326)

Auto-generated PR #326 re-proposed renaming `/fewer-permission-prompts` to `/less-permission-prompts`, a change already investigated and reverted in v1.24.8. The current CHANGELOG mentions neither name and the live skill is `/fewer-permission-prompts`, so the existing content is correct and unchanged.

Arabic parity for the Explore agent model correction

The Arabic subagents module, subagents and skills quizzes, reference cheatsheet, terminal simulator, and feature catalog mirror the Explore-inherits-the-session-model correction.

advanced-features: document fallback model chains

Added a Fallback Model Chains section covering the `--fallback-model` flag and the `fallbackModel` settings array: tried in order when the primary model is overloaded, unavailable, or returns a non-retryable server error; auth/billing/rate-limit/size/transport errors never switch; capped at three models; the switch lasts the current turn only; the flag takes precedence over the setting (code.claude.com/docs/en/model-config).

Reference cheatsheet and feature catalog: --fallback-model entry

Added the `--fallback-model <list>` CLI flag to the reference cheatsheet and the feature catalog so the search palette surfaces fallback model chains.

advanced-features quiz: new question on fallback behavior

Added quiz q30 testing what happens when the primary model returns an overloaded error mid-turn with a fallback chain configured (switch for the current turn only, then retry the primary).

Arabic parity for fallback model chains

Mirrored the Fallback Model Chains section, the cheatsheet and feature-catalog entries, and quiz q30 in the Arabic advanced-features module and data files.

memory: CLAUDE.md/CLAUDE.local.md above the cwd load in full at launch

Clarified that `CLAUDE.md` and `CLAUDE.local.md` files in the directory hierarchy above the working directory are loaded in full at launch, while subdirectory files load on demand when Claude reads files in those directories (code.claude.com/docs/en/memory).

getting-started: qualify the Windows PowerShell-tool behavior

Corrected an over-broad claim that PowerShell is the default shell tool on all Windows and that Git for Windows is no longer required. Git Bash is used when Git for Windows is installed; the PowerShell tool is the shell only when Git Bash is absent; with Git Bash present it is rolling out progressively (opt in with CLAUDE_CODE_USE_POWERSHELL_TOOL=1) and is default-on only for Bedrock, Vertex, and Foundry users (code.claude.com/docs/en/setup).

advanced-features: remove unsupported cleanupPeriodDays cleanup scope

The Arabic advanced-features module and quiz claimed cleanupPeriodDays also removes ~/.claude/tasks/, shell-snapshots/, and backups/. The official settings docs list only session files and orphaned subagent worktrees, so the extra scope was removed to match the English module and the docs. The auto-generated PR that tried to add the same claim to the English module was not merged.

Arabic parity for the memory, getting-started, and cleanupPeriodDays fixes

The Arabic memory and getting-started modules mirror the English wording changes, and the Arabic advanced-features module and quiz q7 were corrected back to the documented cleanupPeriodDays scope (session files plus orphaned worktrees).

mcp: document the claude mcp login / logout CLI authentication commands

`claude mcp login <name>` runs a configured server's OAuth flow from the shell without opening the in-session `/mcp` panel (from v2.1.186), and `claude mcp logout <name>` clears its stored credentials. `--no-browser` prints the authorization URL instead of opening a browser — useful over SSH, where you open the URL locally and paste the redirect URL back at the prompt. Added to the mcp module, reference cheatsheet, and feature catalog.

terminal: add claude mcp login and claude mcp logout to the simulator

The terminal simulator now responds to `claude mcp login` (with the --no-browser flag and an OAuth success output) and `claude mcp logout`, and the `/mcp` panel output lists `claude mcp login` under its manage-servers hint.

Arabic parity for the MCP CLI authentication commands

The Arabic mcp module, reference cheatsheet, terminal simulator, and feature catalog mirror the `claude mcp login` / `claude mcp logout` documentation, including the --no-browser / SSH note.

memory: clarify path-scoped rule activation and add a 'Write effective instructions' section

Path-scoped rules in .claude/rules/ trigger when Claude reads files matching the pattern, not on every tool use; rules without a paths field load at launch with the same priority as .claude/CLAUDE.md; circular symlinks are detected and handled gracefully. A new 'Write effective instructions' section covers the context-window visualization and the under-200-lines-per-CLAUDE.md target. Added to the memory module.

mcp, plugins, subagents, workflows: document four confirmed doc nuances

Channels lead with their research-preview status and v2.1.80 minimum (mcp); a single-skill plugin can place SKILL.md at the plugin root with the frontmatter name as the invocation name (plugins); the built-in Explore and Plan agents skip CLAUDE.md and git status while every other subagent loads both (subagents); and dynamic-workflow resume only works within the same Claude Code session, restarting fresh after you exit (workflows).

memory: restore the deleted guided-terminal marker from audit PR #315

Auto-generated PR #315 replaced the English memory module's `<!-- terminal: guided -->` marker with its new paragraph, which would have broken the guided terminal step and left EN out of parity with AR (which kept the marker). The marker is restored and the new paragraph now sits before it, matching the Arabic structure.

Arabic parity across all five module updates

The Arabic memory, mcp, plugins, subagents, and workflows modules mirror every English change — path-scoped rule behavior and the effective-instructions section, the Channels rewording, the single-skill SKILL.md note, the Explore/Plan skip note, and the workflow resume clarification.

mcp: distinguish the server startup timeout from the per-server tool-execution timeout

MCP_TIMEOUT sets the server startup/connection timeout at launch. Tool execution has a separate cap: a `timeout` field in milliseconds on a server's `.mcp.json` entry (e.g. "timeout": 600000 for ten minutes) limits how long any single tool call may run, overrides MCP_TOOL_TIMEOUT for that server only, is a hard wall-clock limit that progress notifications don't extend, and ignores values below 1000. Added to the mcp module.

mcp: reject the imprecise MCP_TIMEOUT re-wording from audit PR #312

Audit PR #312 wanted to relabel MCP_TIMEOUT as 'a per-server startup timeout, not a general per-server timeout'. The docs call it the 'MCP server startup timeout' — a launch-time environment variable, not a per-server field — so the awkward clause was dropped in favor of the accurate startup-vs-tool-execution split.

Arabic parity for the MCP timeout distinction

The Arabic mcp module mirrors the startup-timeout (MCP_TIMEOUT) versus tool-execution-timeout (per-server `timeout` field / MCP_TOOL_TIMEOUT) split, matching the English change.

plugins: document claude plugin prune accurately + the plugin details inventory (PR #309)

claude plugin prune (new in v2.1.121, aliased autoremove) removes auto-installed plugin dependencies that no other installed plugin still requires — directly-installed plugins are never touched, and 'uninstall <plugin> --prune' does both in one step. claude plugin details now lists the component inventory grouped as Skills, Agents, Hooks, MCP servers, and LSP servers with a projected per-session token cost. Added to the plugins module, the reference cheatsheet, and the feature catalog, EN and AR.

mcp: keep ENABLE_TOOL_SEARCH=auto:N (rejected audit PR #308)

An audit PR removed the auto:N variant of ENABLE_TOOL_SEARCH. The MCP reference documents it verbatim — 'auto:N: Threshold mode with a custom percentage, where N is 0-100' (e.g. auto:5 for 5%) — so the variant stays. The mcp module and quiz were left unchanged.

Arabic parity for the plugin prune / details updates

The Arabic plugins module mirrors the corrected claude plugin prune description and the expanded claude plugin details inventory, matching the English changes.

slash-commands: /code-review is the renamed /simplify, not a brand-new command (PR #306)

The '(new in v2.1.147)' tag implied /code-review was new; the changelog records /simplify being renamed to /code-review in v2.1.147. The module and quiz now describe /code-review as the bundled skill formerly called /simplify, in EN and AR.

getting-started: keep Console in the account requirement list (corrects rejected audit PR #304)

An audit PR dropped 'Console' from the supported account list. The install doc states 'Claude Code requires a Pro, Max, Team, Enterprise, or Console account' — Console stays. We adopted the docs' clearer 'requires … account' phrasing while keeping all five account types.

mcp: keep 'claude mcp serve' / 'add-from-claude-desktop' and the v2.1.121 startup-retry version (rejected audit PR #305)

An audit PR removed two documented commands and re-dated the startup-retry behavior to v2.1.187. The MCP reference documents both 'claude mcp add-from-claude-desktop' and 'claude mcp serve', and dates the up-to-three startup retries on transient errors to v2.1.121 (v2.1.187 is the unrelated 5-minute remote-tool idle timeout). The original mcp content was correct and was kept unchanged.

Arabic parity for the /code-review and account-list updates

The Arabic slash-commands module and quiz mirror the /code-review-formerly-/simplify clarification, and the Arabic getting-started module keeps Console in the account list with the 'requires' phrasing.

slash-commands: the command is /fewer-permission-prompts, not /less-permission-prompts (corrects v1.24.5)

v1.24.5 renamed the command the wrong way round. The official name — per the v2.1.105–v2.1.113 release notes and the live /fewer-permission-prompts skill — is /fewer-permission-prompts. Reverted across the slash-commands and skills modules, the reference cheatsheet, and the feature catalog (entry id and name), so search and every surface now show the correct command.

subagents: priority is managed > CLI flag > project > user > plugin (PR #302)

The subagent definition priority no longer lists 'built-in' as a priority level. Per the features-overview docs, built-in subagents are always available and are not part of the name-override priority system, which runs managed > CLI flag > project > user > plugin.

workflows: ultracode trigger rename stays v2.1.160 (rejected audit PR #300)

An audit proposed moving the rename from v2.1.160 to v2.1.154, citing the GitHub CHANGELOG.md. The official workflows docs (code.claude.com/docs/en/workflows) state 'Before v2.1.160 the literal trigger keyword was workflow' and separately note workflows require v2.1.154+ — so v2.1.160 is the documented rename version and the existing content was kept unchanged.

Arabic parity for the command-name and priority fixes

The Arabic slash-commands, skills, and subagents modules, plus the Arabic feature catalog and reference cheatsheet, mirror the /fewer-permission-prompts revert and the subagent-priority correction.

mcp: document the full ENABLE_TOOL_SEARCH value table

The MCP module now spells out the override values from the docs: true forces tool search always on, false loads every definition upfront on every turn, and auto (or auto:N) activates tool search only when tool definitions exceed 10% (or N%) of the context window.

mcp: tool search is enabled by default — reject audit PR #298

An audit PR claimed tool search only works after setting ENABLE_TOOL_SEARCH=auto and that the default loads every MCP definition upfront. The official tool search reference states tool search is enabled by default (definitions deferred and discovered on demand), off by default only on Vertex AI or non-first-party ANTHROPIC_BASE_URL proxies. The original content was correct; PR #298 was a regression and is closed.

mcp quiz: q6 reframed to scope the 10% threshold to auto mode

Quiz q6 previously framed the 10% threshold as automatic default behavior. Per the docs, the 10% threshold is specifically the ENABLE_TOOL_SEARCH=auto mode — under the default (unset), tool search is already on regardless of size. The question now asks what auto does when definitions cross 10%, with explanations corrected to match.

Arabic parity for the MCP tool search correction

The Arabic MCP module and quiz get the same correction: tool search on by default, the full ENABLE_TOOL_SEARCH value table, and q6 reframed around auto mode.

memory: practical guidance on which memory to use (issue #293)

A reader said the memory hierarchy was confusing without context on why it matters. The Memory Hierarchy section now has an 'In practice' callout: use project memory for what a teammate needs to understand the codebase (setup, testing, architecture), user memory for how you personally like to work, and CLAUDE.local.md for entries that only matter to you.

plugins: claude plugin tag is v2.1.118, not v2.1.121

The plugins module attributed claude plugin tag to v2.1.121. The official changelog shows v2.1.118 added claude plugin tag (release git tags with version validation) and v2.1.121 added the separate claude plugin prune command, so the annotation now reads v2.1.118.

subagents: effort keeps max (rejected audit PR #296)

An audit PR removed max from the subagents effort values, claiming v2.1.72 dropped it. The official sub-agents reference lists the effort options verbatim as low, medium, high, xhigh, max, so the existing content is correct and max was kept.

Arabic parity for all three changes

The Arabic memory module gets the same 'In practice' callout, the Arabic plugins module gets the v2.1.118 fix for claude plugin tag, and the Arabic subagents module keeps max in its effort values.

slash-commands: /less-permission-prompts rename completed everywhere

The command is now /less-permission-prompts across the slash-commands module (PR #291), the feature catalog, and the reference cheatsheet. The catalog and cheatsheet had been missed by PR #291 and still showed the old /fewer-permission-prompts name, so the command search index now matches the official spelling. Verified against the changelog (added in v2.1.112).

skills: budget stays at 1% (rejected audit PR #290)

PR #290 repeated the already-rejected #288 proposal to change the skill-description budget from 1% to 2%. The official /en/skills docs state 'The budget scales at 1% of the model's context window' and give 0.02 = 2% only as the example for raising it via skillListingBudgetFraction, so the existing 1% content is correct and was kept.

Arabic parity for the command rename

The Arabic feature catalog and reference cheatsheet now use /less-permission-prompts, matching the English content and the slash-commands module.

subagents: Agent Teams reflects the v2.1.178 tool removal

The existing Agent Teams paragraph now notes that v2.1.178 removed the TeamCreate and TeamDelete tools — with CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 set, every session already has one implicit team, so teammates are spawned directly through the Agent tool's name parameter with no setup step and cleanup happens automatically on session exit. Verified verbatim against the changelog.

skills: budget stays at 1% (rejected audit PR #288)

The skills module and quiz keep '1% of the context window' as the default skill-description budget. The official skills docs say 1% verbatim; the 0.02 = 2% value the audit cited is the example for raising the budget with skillListingBudgetFraction, not the default, and the claimed changelog line does not exist.

advanced-features: ultracode stays a v2.1.160 rename (rejected audit PR #286)

The dynamic-workflow paragraph keeps its accurate wording: the trigger keyword was renamed from workflow to ultracode in v2.1.160. The changelog says 'Renamed the dynamic-workflow trigger keyword from workflow to ultracode', so the audit's 'two separate changes in v2.1.178, not a rename' reframing was incorrect and was not applied.

Arabic parity for the Agent Teams update

The Arabic subagents module mirrors the v2.1.178 Agent Teams change (TeamCreate/TeamDelete removed, implicit team per session, spawn via the Agent tool's name parameter, automatic cleanup on session exit).

hooks: exact count of 30 hook events

The hooks module said Claude Code 'supports 30+ hook events'; the official hooks docs enumerate exactly 30, so the prose now states '30 hook events' for precision.

advanced-features: Auto Mode block thresholds documented

Added the Auto Mode fallback thresholds (3 consecutive blocks or 20 total blocks pause auto mode and resume prompting; not configurable) to the classifier-precedence paragraph, matching the permission-modes docs and making the existing quiz q2 answerable from the module.

Arabic parity for both fact-checks

The Arabic hooks and advanced-features modules mirror both changes — '30 hook event' and the Auto Mode block-threshold sentence.

getting-started quiz: accurate Console wording

Quiz q3 asks 'Which subscription plans include Claude Code access?', so the correct answer stays the four subscription plans (Pro, Max, Team, Enterprise) and the explanation now frames the Anthropic Console accurately as a non-subscription option — API access billed from pre-paid credits, with the same browser sign-in — instead of PR #281's inaccurate 'Console uses API key authentication rather than a subscription' framing. Cloud providers (Bedrock, Vertex AI) are noted too.

Arabic parity for the Console answer fix

The Arabic getting-started quiz q3 mirrors the corrected wording: the four subscription plans remain the answer, and the Console is described as a non-subscription option (pre-paid API access, same browser sign-in).

Skills quiz: removed an unverifiable ${CLAUDE_EFFORT} value

Quiz q6 described `${CLAUDE_EFFORT}` as resolving to 'low, medium, high, xhigh, max, or ultra — the stored value for ultracode'. The docs do not define an 'ultra' effort value for this variable, so the explanation now lists the verified levels low/medium/high/xhigh/max.

Reference cheat sheet: corrected the CLAUDE_EFFORT description

The `CLAUDE_EFFORT` environment-variable entry no longer claims it can resolve to 'ultra for ultracode', matching the feature catalog's accurate list of low/medium/high/xhigh/max.

Arabic parity for the effort-level fix

The Arabic skills quiz q6 mirrors the corrected `${CLAUDE_EFFORT}` level list (low/medium/high/xhigh/max).

/cd in the terminal simulator, cheat sheet, and catalog

The terminal playground simulates `/cd <directory>` (moving the session without breaking the prompt cache) and lists it in `/help`; the reference cheat sheet and feature catalog both gained a `/cd` entry.

--safe-mode and CLAUDE_CODE_SAFE_MODE reference coverage

The cheat sheet documents the `--safe-mode` flag and `CLAUDE_CODE_SAFE_MODE` env var (start with all customizations disabled for troubleshooting), and the feature catalog adds a `--safe-mode` flag entry.

Bundled-skills and model/footer settings in the cheat sheet

Added `CLAUDE_CODE_DISABLE_BUNDLED_SKILLS` / `disableBundledSkills`, plus the `enforceAvailableModels` (v2.1.175) and `footerLinksRegexes` (v2.1.176) managed settings, to the reference cheat sheet's env-vars and config-options sections.

Arabic parity for the new coverage

The Arabic terminal commands, cheat sheet (slash commands + CLI flags), and feature catalog mirror the `/cd` and `--safe-mode` additions.

MCP reconnection now covers SSE servers and startup retries

The MCP module states that HTTP or SSE servers reconnect mid-session with exponential backoff (up to five attempts), and that as of v2.1.121 the initial connection at startup is retried up to three times on transient errors such as a 5xx response, connection refused, or timeout.

Channels note the v2.1.80 minimum version

The Channels section now states that the Telegram, Discord, and iMessage channel plugins require Claude Code v2.1.80 or later, matching the official channels documentation.

Removed a duplicate MCP quiz question

The MCP quiz had two near-identical questions both testing the `alwaysLoad` config option; the redundant one is removed and the remaining questions renumbered to q1–q9.

Corrected the Channels feature-catalog entry

The catalog described channels as 'Telegram, Slack, and other platforms', and the `channelsEnabled` managed setting likewise listed 'Telegram, Slack'; both now correctly list the Telegram, Discord, and iMessage channel plugins, and the feature entry notes the research-preview / v2.1.80 requirement. Slack is a separate integration, not a channel.

Arabic parity for the MCP fixes

The Arabic MCP module mirrors the SSE/startup-retry and v2.1.80 channel updates, the Arabic MCP quiz is renumbered to a matching q1–q9, and the Arabic feature catalog's Channels entry is corrected.

Corrected the ExitWorktree availability claim for isolated subagents

The Subagents module no longer says `isolation: worktree` agents can use `ExitWorktree` to leave their worktree early — the tools reference states `ExitWorktree` is not available to subagents that run in their own working directory. The passage now covers worktree locking during the run and the `worktree.baseRef` base-branch behavior instead.

Fixed the worktree.baseRef default in Advanced Features

The module now lists `fresh` (branch from the remote default for a clean tree) as the default for `worktree.baseRef`, with `head` as the opt-in for branching from local HEAD — matching the official docs, the module quiz, and the reference cheatsheet.

Arabic parity for both worktree fixes

The Arabic Subagents and Advanced Features pages mirror the ExitWorktree and baseRef corrections.

Removed fictional deniedPlugins managed-policy setting

The Plugins module now lists only the managed-policy settings that exist in the official docs — `enabledPlugins`, `extraKnownMarketplaces`, `strictKnownMarketplaces`, and `blockedMarketplaces` — instead of the non-existent `deniedPlugins`.

Plugins quiz distractor corrected

The team-disable quiz question no longer presents `deniedPlugins` as a real setting; the distractor and its explanation now turn on the real `blockedMarketplaces` managed-policy blocklist.

Arabic parity for the deniedPlugins fix

The Arabic Plugins module and quiz mirror the English correction, dropping `deniedPlugins` for the real managed-policy settings.

Settings precedence wording made unambiguous

The Project Setup module now states plainly that project settings are overridden by local settings, and that only managed settings and command-line arguments override local — matching the official precedence order (managed > command-line > local > project > user).

Arabic parity + managed-settings terminology consistency

The Arabic Project Setup page mirrors the precedence clarification and now uses 'الإعدادات المُدارة' consistently for the managed layer instead of switching to 'سياسة المنظمة'.

/ultrareview free runs: no fixed May 5, 2026 expiry

The Workflows module no longer claims the 3 Pro/Max free /ultrareview runs expire on May 5, 2026. Per the official ultrareview docs, the three runs are a one-time allotment per account that does not refresh — there is no published expiry date. Corrected in EN and AR.

slash-commands quiz: /code-review vs /simplify wording aligned with docs

Quiz q7 keeps the doc-accurate 'renamed from /simplify in v2.1.147' framing (matching the commands quiz, reference cheatsheet, and templates), and its closing note now matches the docs: from v2.1.154 /simplify is a separate cleanup-only review that applies fixes without hunting for bugs, and /code-review --fix is the bug-finding-with-fixes path.

Arabic parity for both fixes

The Arabic Workflows module and slash-commands quiz mirror the /ultrareview free-runs correction and the /code-review wording, with natural phrasing in place of an earlier mixed-script glitch.

Skill description budget is ~1% of the context window by default

Skills module and quiz corrected from ~2% to ~1%, the documented default. The 2% figure is only an example of a raised budget, not the default. Raise it with the skillListingBudgetFraction setting or the SLASH_COMMAND_TOOL_CHAR_BUDGET environment variable. Verified against the official skills docs.

Use /doctor (not /context) to check the skill budget

The Skills module and quiz now point to /doctor to see whether the skill-description budget is overflowing and which skills are dropped from the listing, matching the official docs.

Arabic parity and tanween fix

The Arabic Skills module and quiz mirror the 1%/budget corrections, and a tanween spelling fix (مثلاً → مثلًا) is applied in the budget example.

/terminal-setup no longer described as a notification configurator

Getting Started now explains that `/terminal-setup` writes the Shift+Enter keybinding for VS Code, Cursor, Devin Desktop, Alacritty, and Zed — not notification support for Kitty/Ghostty/Alacritty. Desktop notifications work by default in Ghostty, Kitty, and iTerm2; other terminals use `preferredNotifChannel: "terminal_bell"` or a Notification hook. Verified against the terminal-config docs page.

Restored the GPU-acceleration and scroll-sensitivity detail for VS Code forks

In VS Code, Cursor, and Devin Desktop, `/terminal-setup` turns off the integrated terminal's GPU acceleration (to prevent garbled text) and adjusts its scroll sensitivity for smoother fullscreen scrolling. This accurate detail was preserved and made precise rather than dropped. The feature-catalog description was sharpened to match.

Arabic parity for the /terminal-setup correction

The Arabic Getting Started module and feature catalog mirror the English corrections with consistent phrasing for the Shift+Enter keybinding scope, the notification defaults, and the GPU-acceleration/scroll-sensitivity behavior.

⌘K command palette for site-wide search

A new search island lets you open a command palette with ⌘K / Ctrl+K and find modules, commands, and reference material from any page, with keyboard navigation through results.

/code-review documents the ultra effort level

Commands module, slash-commands module, quizzes, reference cheatsheet, and terminal simulation now list the full documented effort range (low/medium/high/xhigh/max/ultra). `/code-review ultra` runs a deeper multi-agent review in the cloud — the same one previously surfaced as `/ultrareview`.

/simplify is no longer described as an alias for /code-review

From v2.1.154 /simplify runs a separate cleanup-only review that applies fixes without hunting for bugs. The commands module, slash-commands module, both quizzes, reference cheatsheet, templates, terminal-steps, and terminal simulation all drop the 'still works as an alias' framing and now explain the v2.1.154 split, with /code-review as the bug-finding tool.

Dynamic workflow trigger keyword renamed to ultracode

The workflows module, workflows quiz (Q9), reference cheatsheet /workflows entry, and terminal simulation now describe the trigger as `ultracode` instead of `workflow`. The rename landed in v2.1.160; asking for a workflow in your own words still works. Verified against the workflows docs page.

MAX_THINKING_TOKENS=0 scoped to Opus 4.6 and Sonnet 4.6

Advanced-features module clarifies that `MAX_THINKING_TOKENS=0` only disables thinking on Opus 4.6 and Sonnet 4.6. On Opus 4.7 and later, adaptive reasoning is always used and the variable has no effect. Verified against the model-config docs page.

/install-github-app: drop unverified sandboxed-environment claim

The workflows module no longer states that `@claude` mentions run in a sandboxed environment — the github-actions docs say `runs-on: ubuntu-latest` and 'Your code stays on GitHub's runners' but do not use the 'sandboxed environment' framing.

Arabic parity for all five changes

Arabic versions of the commands, slash-commands, workflows, and advanced-features modules, plus the Arabic quizzes, reference cheatsheet, and terminal simulation all mirror the EN updates with consistent phrasing for the effort-range extension, the /simplify split, the ultracode trigger rename, and the MAX_THINKING_TOKENS scoping.

waitingFor field disambiguates what a waiting session needs

When `status` is `waiting`, the new `waitingFor` field (v2.1.162) says exactly what the session is blocked on — `permission prompt` or `input needed`. The subagents module, templates, terminal simulation, and reference cheatsheet now include a jq example that wakes only permission-prompt sessions with `claude respawn`, so a boot script can route those two cases to different handlers.

OTEL_RESOURCE_ATTRIBUTES env var documented

Advanced-features module now documents `OTEL_RESOURCE_ATTRIBUTES` as the supported way to attach team, department, or cost-center labels to every metric datapoint and event. Formatting rules (comma-separated, no spaces, percent-encode special characters) and the `OTEL_METRICS_INCLUDE_RESOURCE_ATTRIBUTES=false` cardinality escape hatch are called out. Built-in attributes like `user.id` and `session.id` always win on key collision.

claude agents --json field names match the docs

Subagents module, terminal simulation, templates, quiz Q9 explanation, reference cheatsheet, and feature catalog now list the documented `pid`, `cwd`, `kind`, `startedAt`, `sessionId`, `name`, and `status` fields — replacing the older `id`/`state`/`project`/`last_activity` names. All jq examples are updated from `.state` to `.status` and from `.id` to `.sessionId` so they actually parse the live output.

Arabic parity for all three changes

Arabic subagents module, advanced-features module, terminal simulation, templates, quiz, reference cheatsheet, and feature catalog all mirror the field-name correction, the `waitingFor` addition, and the `OTEL_RESOURCE_ATTRIBUTES` env var, with consistent Arabic phrasing for the OTLP resource block and the cardinality escape hatch.

Kiro listed as a supported VS Code fork

The official VS Code extension docs page now states the extension installs in 'other VS Code forks like Devin Desktop or Kiro'. The getting-started module and Q5 quiz explanation in the getting-started quiz both add Kiro alongside Cursor and Devin Desktop so learners see the full set of VS Code forks where the Claude Code extension is supported.

Windsurf is now Devin Desktop in the Claude Code UI

Getting-started module no longer presents Windsurf as a current VS Code fork name — it now describes Cursor, Devin Desktop (renamed from Windsurf in v2.1.162), and Kiro as the VS Code forks that support the Claude Code VS Code extension and the integrated terminal. The paragraph also notes that `/ide` and `/scroll-speed` use the new Devin Desktop label, on top of the existing `/terminal-setup` auto-configuration.

Arabic parity for the Devin Desktop rename and Kiro addition

The Arabic getting-started module and Arabic getting-started quiz mirror the rename — Cursor وDevin Desktop (الاسم الجديد لـ Windsurf من v2.1.162) وKiro — with consistent phrasing for the `/ide`, `/terminal-setup`, and `/scroll-speed` label change.

Auto Mode availability on Bedrock, Vertex AI, and Foundry

Advanced-features module now documents Auto Mode platform availability: on the Anthropic API it's on by default once your account meets the model and admin requirements; on Bedrock, Vertex AI, and Foundry (v2.1.158+) it's gated behind `CLAUDE_CODE_ENABLE_AUTO_MODE=1` and only Opus 4.7 and Opus 4.8 are supported, while Sonnet 4.5, Opus 4.5, Haiku, and Claude 3 are not. The `permissions.disableAutoMode: "disable"` admin override is called out as the way to lock the mode off in managed settings.

Workflows module: drop the unsupported `/ultrareview` restriction hedge

The /ultrareview paragraph used to caveat that workflows generally work on Bedrock, Vertex AI, and Foundry while `/ultrareview` specifically might be restricted. The official workflows page now states plainly that dynamic workflows are available on all paid plans on Bedrock, Vertex AI, and Foundry with Claude Code v2.1.154+, so the hedge is removed in favor of the confirmed availability.

Arabic parity for both Auto Mode and workflows updates

Both changes are mirrored in the Arabic advanced-features and workflows modules with consistent phrasing for platform availability, the `CLAUDE_CODE_ENABLE_AUTO_MODE` opt-in, and the model restriction to Opus 4.7/4.8 on third-party providers.

Dynamic-workflow trigger renamed to `ultracode`

Advanced-features and commands modules now document the v2.1.160 rename of the trigger keyword from `workflow` to `ultracode`. The literal `workflow` keyword no longer triggers a run, but a natural-language request still works, and `/effort ultracode` activates dynamic workflows.

CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE removed (no-op)

Removed all instructions to set this env var: from the commands module fast-mode section, the advanced-features environment-variable section and templates, and the feature catalog. Quiz Q23 in advanced-features now teaches that the variable was removed in v2.1.160 with no replacement.

/ultrareview introduced in v2.1.111

Workflows module now reflects the official CHANGELOG — /ultrareview has been available since Claude Code v2.1.111, instead of being marked only as new in v2.1.112. (This entry originally read v2.1.86; corrected in v1.24.31.)

Arabic parity across all three audit fixes

Every change is mirrored in the Arabic modules (advanced-features, commands, workflows), Arabic templates, Arabic feature catalog, and Arabic quizzes, with Arabic phrasing for the `workflow`/`ultracode` distinction clarified after CodeRabbit review.

Fast mode covers Opus 4.8, 4.7, and 4.6

Commands module no longer pins fast mode to Opus 4.6 alone. Opus 4.8 is the fast mode default in v2.1.154+, and the CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE=1 env var is documented for pinning back to 4.6 while it remains supported. The Opus 4.6 fast mode deprecation timeline is called out.

Fast mode platform exclusions are complete

Commands module now lists all platforms where fast mode is unavailable — Bedrock, Vertex AI, Foundry, Claude Platform on AWS — and notes it is not supported in the VS Code extension. The rate-limit pool is shared across Opus 4.8, 4.7, and 4.6.

Project-setup quiz Q4 precedence explanation

Project-setup quiz Q4 option B explanation was inverted: settings.local.json (not project settings) overrides project. The corrected explanation now reads: project settings override user settings, but they are overridden by local settings, managed policy, and command-line arguments — matching the official settings hierarchy.

Arabic parity for fast mode and project-setup quiz

Every change is mirrored in the Arabic commands module and the Arabic project-setup quiz, with consistent terminology for deprecation and platform exclusions.

Plugin monitors nest under experimental

Plugins module now shows monitors declared under experimental.monitors in plugin.json, with a note that top-level still works but claude plugin validate warns. Feature catalog entry updated to match.

Project setup mentions monitors/themes are experimental

Project-setup module now tells learners that monitors and themes are experimental plugin manifest keys and should be nested under experimental: {}.

Arabic parity for plugin manifest guidance

Mirrored in the Arabic plugins and project-setup modules plus the AR feature catalog, with cleaner Arabic phrasing for the new project-setup paragraph.

IDE extensions: Cursor and Windsurf are VS Code forks

Getting-started module now describes Cursor and Windsurf as VS Code forks that reuse the VS Code extension and integrated terminal, with /terminal-setup auto-configuring GPU acceleration and scroll sensitivity. Quiz updated to match.

Subagents have their own auto memory

Memory module now notes that subagents can maintain their own auto memory, with a link back to the subagents module for configuration details.

DISABLE_UPDATES vs DISABLE_AUTOUPDATER

Project-setup module now distinguishes DISABLE_UPDATES=1 (blocks all update paths, including manual claude update) from DISABLE_AUTOUPDATER (suppresses package-manager update notifications only).

Skill description budget is ~2% of the context window

Skills module and quiz updated from the older ~1% figure to the current ~2% budget, with SLASH_COMMAND_TOOL_CHAR_BUDGET callable to raise the cap. All q3 option explanations now agree.

/simplify and /code-review behavior

Slash-commands module clarifies that /code-review is a bundled skill that reports findings, /code-review --fix applies them to the working tree, and /simplify now invokes /code-review (report only) — the old cleanup-and-fix /simplify behavior was removed, but /simplify still works as an alias.

CLAUDE_CODE_FORK_SUBAGENT scope

Subagents module and quiz now state that forked subagents work in interactive sessions, non-interactive mode, the Agent SDK, and on external builds — dropping the inaccurate 'interactive mode only' framing.

/ultrareview platform availability

Workflows module no longer claims /ultrareview is unavailable on Bedrock, Vertex AI, or Foundry. Learners are pointed to the official docs to confirm, since the workflows feature itself works on those platforms.

Arabic parity for all fixes

Every fix is mirrored in the Arabic module pages and quizzes, including the cross-module link in memory.md and the consistent 2% phrasing across all skills q3 explanations.

Clarified the channels research-preview wording

MCP module now says the Telegram, Discord, and iMessage channel plugins are 'included in' the research preview, matching the official docs phrasing.

Added dynamic workflows

Workflows module now covers dynamic workflows (v2.1.154+) — orchestrating dozens to hundreds of subagents from a script Claude writes, the bundled /deep-research workflow, /effort ultracode, saving a run as a /command, and disabling via /config or CLAUDE_CODE_DISABLE_WORKFLOWS=1. Mirrored in the feature catalog, reference cheatsheet, and terminal simulation (/workflows, /deep-research).

Expanded the built-in skills list

Skills module now lists the full set of bundled skills available in every session — /code-review, /batch, /debug, /loop, and /claude-api — alongside the /run, /verify, and /run-skill-generator trio (v2.1.145+). Added /claude-api to the terminal simulation.

New quiz coverage

Added a workflows quiz question on how to start a dynamic workflow (the 'workflow' keyword or /effort ultracode, not /workflows) and a skills quiz question on bundled skills. Corrected the skills effort-variable answer to include the ultra value.

Arabic parity for all changes

Translated the dynamic workflows section, bundled-skills table, new quiz questions, feature-catalog entries, and cheatsheet rows into Arabic.

Corrected reasoning effort levels

Hooks and subagents modules now list the full effort set — low, medium, high, xhigh, max, auto — matching official docs. xhigh was added for Opus 4.7, sitting between high and max.

Clarified the PowerShell tool shell setting

Skills module now notes that enabling the PowerShell tool with CLAUDE_CODE_USE_POWERSHELL_TOOL=1 on Linux and macOS also requires pwsh on your PATH.

Fixed Arabic permission-mode cycle order

Commands module (Arabic) now shows Shift+Tab cycling default → acceptEdits → plan, matching the English module and official docs.

Documented the cloud code review

Slash commands module, feature catalog, reference cheatsheet, and terminal simulation now cover /code-review ultra (alias /ultrareview) — a deep, multi-agent code review run in a cloud sandbox.

Localized feature catalog categories

Arabic feature catalog now uses Arabic category labels for the new entries, and fenced code blocks carry language tags.

Added session recap and command history

Commands module now covers automatic session recaps when returning to the terminal, /recap on-demand summaries, and Ctrl+R reverse history search with Ctrl+S scope cycling.

Added mobile push notifications

Workflows module now documents push notifications via Remote Control — Claude sends phone alerts when tasks finish or need input.

Expanded auto mode configuration

Advanced features module now documents hard_deny rules (unconditional blocks), four-tier precedence, and claude auto-mode defaults/config/critique CLI subcommands.

Added new slash commands

Slash commands module now includes /recap, /fewer-permission-prompts, /tui, /focus, and /undo (alias for /rewind).

Added hook configuration options

Hooks module now documents continueOnBlock (feed rejection back to Claude instead of ending the turn) and terminalSequence (emit notifications without a terminal).

Full EN/AR parity for all changes

All new content mirrored in Arabic modules, quizzes, feature catalog, and reference cheatsheet.

Added output styles section

Commands module now covers built-in output styles (Default, Proactive, Explanatory, Learning) and custom output style creation with Markdown files and keep-coding-instructions frontmatter.

Added channels section

MCP module now documents channels — MCP servers that push events (Telegram, Discord, iMessage) into running sessions. Covers --channels flag, pairing, allowlists, and enterprise controls.

Added computer use section

Advanced features module now covers the computer-use MCP server for GUI control on macOS. Includes enabling, permissions, per-app approval, and safety features.

Expanded routines documentation

Workflows module now fully documents routines: API triggers, GitHub triggers, web UI at claude.ai/code/routines, connectors, one-off scheduled runs, and /schedule list/update/run CLI management.

Full EN/AR parity for all changes

All new content mirrored in Arabic modules, quizzes, feature catalog, and reference cheatsheet.

Fixed auto-update documentation accuracy

Project-setup module now correctly documents that Homebrew/WinGet can opt in to auto-updates via CLAUDE_CODE_PACKAGE_MANAGER_AUTO_UPDATE=1, adds autoUpdatesChannel setting, and DISABLE_AUTOUPDATER env var. Previous PR #195 incorrectly claimed the env var was unsupported.

Removed incorrect npm deprecation claim

Getting-started module, quiz, and installation tabs no longer describe npm as deprecated — the official docs list it as a supported alternative installation method.

Added /run, /verify, /run-skill-generator bundled skills

Documents the three bundled skills for launching, verifying, and recording app build recipes. Added to both English and Arabic skills module.

Added skillOverrides setting

Documents the skillOverrides setting for controlling skill visibility from settings.json without editing SKILL.md frontmatter. Includes the /skills interactive menu workflow.

Added string substitution variables

Documents ${CLAUDE_SESSION_ID}, ${CLAUDE_EFFORT}, ${CLAUDE_SKILL_DIR}, and named arguments frontmatter in the skills module.

Added Linux package managers and claude doctor

Getting-started module now documents apt, dnf, and apk installation methods and the claude doctor verification command.

Full EN/AR parity for all changes

All module text, quiz corrections, and tab content mirrored in Arabic.

Fixed CLAUDE_CODE_FORK_SUBAGENT availability and Arabic description

English corrected from 'interactive mode only' to 'interactive sessions, non-interactive mode, and the Agent SDK'. Arabic completely rewritten to match (was incorrectly described as external-builds-only).

#192

Corrected /less-permission-prompts version

Version corrected from v2.1.112 to v2.1.111 (confirmed via CHANGELOG) in both English and Arabic skills module.

#191

Added PostToolUse updatedToolOutput documentation

PostToolUse hooks can replace tool output entirely via hookSpecificOutput.updatedToolOutput (works for all tools since v2.1.121). Quiz q8 tested it, but the module content was missing.

#106

Added --add-dir and --mcp-config CLI flags

--add-dir grants Read/Edit access to additional directories; --mcp-config loads MCP servers from external JSON for the current session. Both documented in subagents module and terminal simulator.

#171

Added missing args exec form to Arabic hooks

The English hooks module documented the exec form (args array for shell-free process spawning) but Arabic was missing it entirely.

Added /code-review bundled slash command

Reviews the current diff for correctness bugs and reports findings without editing files. Accepts effort levels (low/medium/high/xhigh/max) and --comment to post inline GitHub PR comments. Renames /simplify, which still works as an alias.

#184

Full Arabic parity for /code-review

Mirrored across Arabic module content, quizzes, terminal steps, terminal-commands.json, templates, the feature catalog, and the reference cheatsheet.

#184

Added claude agents --json for scripting

claude agents --json prints the session roster as a JSON array — id, state, name, project, last activity — for tmux-resurrect-style boot scripts, status bars, and session pickers.

#179

Added Stop/SubagentStop background_tasks and session_crons fields

Stop and SubagentStop hook inputs now include background_tasks (running background bash/subagents) and session_crons (queued scheduled tasks). A completion-gate hook can block until both arrays drain.

#179

Added /plugin Discover/Browse component preview

The /plugin Discover and Browse screens now show a plugin's commands, agents, skills, hooks, and MCP/LSP servers inline before installation.

#179

Added Read tool partial-view truncation

When a whole-file read exceeds the token limit, Read now returns a truncated first page with a PARTIAL view notice instead of a hard error. Claude paginates with offset/limit or narrows with grep.

#179

Renamed /extra-usage to /usage-credits

v2.1.144 renamed 'extra usage' to 'usage credits' across the CLI; /extra-usage still works as an alias. Catalog, cheatsheet, slash-commands module, fast-mode copy, and terminal sim updated.

#174

Full Arabic parity for v2.1.144 + v2.1.145

All five features mirrored across Arabic module content, quizzes, terminal steps, templates, terminal-commands.json, feature catalog, and reference cheatsheet.

#174, #179

Audit corrections across 4 modules

Fact-check fixes in memory, project-setup, hooks, and plugins modules from automated audit pipeline.

#176, #177, #178, #180, #181

Added claude plugin enable/disable commands

Toggle installed plugins on or off without uninstalling. Accepts --scope (user/project/local) to control where the setting is stored.

#171

Added /plugin marketplace projected context cost

Plugin Discover/Browse screens now show projected token cost per session for each plugin.

#171

Added plugin details and plugin tag commands

claude plugin details <name> shows component inventory and token cost; claude plugin tag creates release git tags with version validation.

#171

Migrated changelog to curated YAML format

Changelog data moved from GitHub API scraping to a structured YAML file (changelog.yaml) with per-release entries, categories, and issue references.

#170

Added agents --cwd directory-filtered Agent view

claude agents --cwd <path> filters the Agent view to only show agents running in that directory.

#168

Added --permission-mode flag with full accepted-modes list

Documents --permission-mode <mode> flag with the complete list of accepted permission modes.

#168

Added ANTHROPIC_WORKSPACE_ID workload identity federation

Documents ANTHROPIC_WORKSPACE_ID for Workload Identity Federation on cloud platforms.

#168

Added CLAUDE_CODE_PLUGIN_PREFER_HTTPS for CI plugin cloning

CLAUDE_CODE_PLUGIN_PREFER_HTTPS=1 ensures plugins are cloned over HTTPS in CI environments.

#168

Brightened sidebar collapse toggle

Higher-contrast text in expanded state with full text-primary color and thicker chevron stroke when collapsed.

#162

Fixed backwards CLAUDE_CODE_NATIVE_CURSOR description

Corrected the pre-existing backwards description for CLAUDE_CODE_NATIVE_CURSOR.

#164

Added 3 publicly documented env vars from v2.1.138/140

CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE, CLAUDE_CODE_RESUME_PROMPT, and CLAUDE_CODE_ENABLE_FEEDBACK_SURVEY_FOR_OTEL across modules, feature catalog, quizzes, and templates.

#164

Added --plugin-url trust warning

Added a security note for --plugin-url in both English and Arabic plugins documentation.

#160

Added 6 verified features across 5 modules

Gateway model discovery, alternate screen toggle, image pasting hint, --plugin-url flag, package manager auto-update, and SESSION_ID env var.

#160

Added 12 quiz questions across 6 modules

New quiz questions covering the 6 verified features added this release, with correct numbering throughout.

#160

Complete Arabic quiz parity across all 7 modules

15 missing Arabic quiz questions added (getting-started q7-q8, hooks q9-q10, project-setup q6, advanced-features q13-q21), plus 4 terminal steps and 4 templates.

#160

Added 7 features from v2.1.138/139

Module content, quizzes, templates, and feature catalog entries for 7 newly verified Claude Code features.

#159

Added 5+ features from v2.1.128–133

Module content, quizzes, and templates for 5+ newly verified Claude Code features.

#142

Corrected Shift+Tab permission mode cycling order

Shift+Tab now correctly cycles through permission modes in the right order instead of skipping states.

#135

New ultrareview --timeout template

Added a CI code review workflow template that runs ultrareview with --timeout for environments where long reviews are disruptive.

#121

Added 7 features from v2.1.119–126

Module content, quizzes, and templates for 7 newly verified Claude Code features.

#119

Added 10 features from v2.1.118–123

Full content, quiz questions, and templates for 10 verified features across commands, plugins, MCP, and advanced-features.

#114, #96

New beginner introduction and glossary

The Getting Started module now includes a step-by-step beginner guide and a glossary of terms for those new to Claude Code.

#95

Added Advisor Tool quiz and terminal step

Full quiz question and terminal step coverage for the Advisor Tool in both English and Arabic.

#94

Complete Getting Started module

Full Module 0 coverage for install, authentication, terminal, and IDE setup in English and Arabic.

#86

Corrected Advisor Tool description

Advisor Tool description is now accurate. Undocumented env vars have been removed from the feature catalog.

#84

Quiz no longer auto-advances after correct answer

Learners can read explanations before deciding to continue. A 'Next Question' button replaces the automatic advance.

#20

Removed /allowed-tools from Configuration commands

/allowed-tools was listed as a built-in command but does not exist in Claude Code. Removed it from the slash-commands module.

#33

Fixed 5 mismatched Arabic quiz questions

Arabic quiz questions in memory, skills, and slash-commands modules now match their English counterparts.

#21

Corrected effort max scoping

The effort maximum is session-wide, not Opus-only. Documentation corrected across 4 modules.

#74

Added sandbox.network.deniedDomains coverage

Documents sandbox.network.deniedDomains in advanced-features with quiz questions and templates.

#71

Added v2.1.105 coverage across 5 modules

Module content updates for commands, hooks, plugins, skills, and subagents covering newly verified Claude Code features.

#58

Corrected Arabic context budget figure

Skill description budget corrected from 2% / 16,000 to 1% / 8,000 characters.

#28

Documented model and shell frontmatter fields

Added documentation for model and shell frontmatter fields in the skills module.

#28

/clear command clears the terminal

The /clear command in the terminal simulator now clears visible history instead of printing text.

#25

Card download shows error on failure

When html2canvas fails, a localized error message appears below the button instead of silently failing.

#25

Template copy button no longer overlaps text

Code template blocks push content below the copy button to prevent text rendering underneath it.

#25

New audit-quiz skill

Automated skill for checking quiz data integrity, structural validity, and EN/AR consistency.

#23

New audit-terminal skill

Automated skill for cross-referencing module content against terminal-commands.json and terminal-steps.yaml files.

#23

New /allowed-tools command

Terminal simulator now supports /allowed-tools for managing permitted tools with add/remove and wildcard support.

#22

Migrated to CSS logical properties

Replaced physical CSS properties with logical equivalents across 8 components for proper RTL support.

#18

RTL-aware keyboard navigation

Added RTL-aware arrow-key navigation to BuildTabs, QuizBlock, and InteractiveDiagram.

#18

Arabic self-assessment scoring corrected

Self-assessment scores now correctly cap at 20. Expert users are no longer misclassified as Beginner.

#18

Feedback form mixed-script input support

Free-text fields use dir='auto' for mixed Arabic/English input. Email fields use dir='ltr'.

#18

Corrected slash-commands and subagents documentation

Fixed command discovery, model aliases, effort options, diagnostics, and Agent Teams documentation.

#15

GLM audit workflow for learning modules

Automated audit workflow with OpenCode agents, runner scripts, report templates, and tracker sync.

#15

Self-assessment scores correctly up to 20 points

Score ranges are Beginner 0–6, Intermediate 7–14, Advanced 15–20. Expert users are no longer classified as Beginner.

#11

Public feedback tracker

New /feedback page showing active feedback organized by status with item counts and a submission form.

#7

Changelog page for shipped items

New /changelog page showing what has been shipped and fixed, linked from the footer.

#7

Report an issue link on every learn page

Each module page has a link that opens the feedback form with the module name pre-filled.

#7

Navigation renamed for clarity

Build → Config Builder, Quick Reference → Cheat Sheet, Feature Catalog → Feature Index. Footer and sidebar reorganized.

#7