From deb7e5c7106cde427e8bd6aefd79bbdf35ffa351 Mon Sep 17 00:00:00 2001 From: Luong NGUYEN Date: Mon, 17 Aug 2026 12:44:44 +0700 Subject: [PATCH] docs: sync to Claude Code v2.1.233 (#168) * docs: sync to Claude Code v2.1.233 Bring the tutorial from the v2.1.220 baseline up to v2.1.233 (13 releases) per update-plan-2026-08-15.md. Corrections: - Remove /ultraplan (removed in v2.1.222); point at plan mode instead - Drop the 200-subagent-per-session cap and CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION (cap removed in v2.1.224) - Note TaskCompleted/TaskCreated only fire when todo tools are enabled Updates: - /review documented as an alias of /code-review (v2.1.223) - fork subagents default to interactive (v2.1.232) - ultrareview v2.1.227 floor, --post/--no-post - GitLab MR support for -w/--worktree and Agent View New sections: - Cross-session messaging (ListAgents, SendMessage, crossSessionInbound) - Sandbox credential masking (v2.1.221, v2.1.224) - Trust and permission scoping - archive and command plugin sources, conditional plugin activation - claude self-hosted-runner, 7 new env vars Translations (vi/ja/zh/uk) mirrored where the corresponding pages exist. Every modified file carries the 2.1.233 metadata footer; files already at 2.1.220 with no content change were left alone. * docs: address review findings from the v2.1.233 sync - Mirror the /review alias correction (v2.1.223) into vi/ja/uk/zh 01-slash-commands: add the live alias row, drop the stale row from the deprecated table, and rewrite the history bullet. All four still called /review deprecated in favour of a code-review plugin while carrying a 2.1.233 stamp. - Rewrite the 10-cli CLAUDE_CODE_FORK_SUBAGENT row to agree with 04-subagents: fork mode is on by default in interactive sessions on every build since v2.1.232; the variable enables it where it is off by default. - Remove a stray blank line that split the 10-cli environment-variable table, leaving the seven new v2.1.221-v2.1.233 rows without a delimiter row and rendering them as literal pipe text. - Add the todo-tools caveat to the TaskCompleted row alongside TaskCreated in claude_concepts_guide, uk/ja concepts guides, and the uk/vi/zh/ja CATALOGs. - Fix the ja 07-plugins and 09-advanced-features footers, which paired a 2.1.233 version stamp with an April 24 last-updated date. * docs(subagents): correct the stale fork-subagent claim in ja pages (#171) ja/10-cli and ja/04-subagents both stated that CLAUDE_CODE_FORK_SUBAGENT only enables forked subagents on external builds (Bedrock, Vertex, Foundry) with no effect on the Anthropic API. That was true at v2.1.117 but has been wrong since v2.1.232: fork mode is on by default in interactive sessions on every build, first-party or not, and the variable now turns it on where it is off by default (claude -p, the Agent SDK, or Claude Code older than v2.1.232). Both pages remain stamped 2.1.119 on purpose. Only the individual claim was verified and corrected; bumping the page stamp would assert a whole-page verification that has not been done. Do not bump these stamps without a full resync of the pages. --- 01-slash-commands/README.md | 30 ++-- 04-subagents/README.md | 23 +-- 06-hooks/README.md | 11 +- 07-plugins/README.md | 96 +++++++++++- 09-advanced-features/README.md | 240 +++++++++++++++++++++--------- 10-cli/README.md | 37 +++-- CATALOG.md | 27 ++-- INDEX.md | 6 +- QUICK_REFERENCE.md | 8 +- README.md | 7 +- claude_concepts_guide.md | 8 +- ja/01-slash-commands/README.md | 20 ++- ja/04-subagents/README.md | 2 +- ja/06-hooks/README.md | 6 +- ja/07-plugins/README.md | 25 +++- ja/09-advanced-features/README.md | 110 ++++---------- ja/10-cli/README.md | 2 +- ja/CATALOG.md | 14 +- ja/INDEX.md | 16 +- ja/README.md | 7 +- ja/claude_concepts_guide.md | 10 +- uk/01-slash-commands/README.md | 16 +- uk/06-hooks/README.md | 8 +- uk/07-plugins/README.md | 24 ++- uk/09-advanced-features/README.md | 10 +- uk/CATALOG.md | 13 +- uk/INDEX.md | 8 +- uk/README.md | 8 +- uk/STYLE_GUIDE.md | 6 +- uk/claude_concepts_guide.md | 12 +- vi/01-slash-commands/README.md | 16 +- vi/06-hooks/README.md | 8 +- vi/07-plugins/README.md | 24 ++- vi/CATALOG.md | 12 +- vi/INDEX.md | 7 +- zh/01-slash-commands/README.md | 13 +- zh/06-hooks/README.md | 9 +- zh/07-plugins/README.md | 25 ++++ zh/CATALOG.md | 13 +- zh/INDEX.md | 7 +- zh/README.md | 8 +- zh/STYLE_GUIDE.md | 7 + 42 files changed, 630 insertions(+), 329 deletions(-) diff --git a/01-slash-commands/README.md b/01-slash-commands/README.md index 5def86c..9f8f1d0 100644 --- a/01-slash-commands/README.md +++ b/01-slash-commands/README.md @@ -25,7 +25,7 @@ Built-in commands are shortcuts for common actions. There are **60+ built-in com | `/add-dir ` | Add working directory | | `/agents` | Manage agent configurations | | `/branch [name]` | Switch into a copy of the conversation at this point, preserving the original (return to it with `/resume`) | -| `/fork [prompt]` | Copy the current conversation into a new **background session** and keep working here; the two are independent from that point on and the copy gets its own row in `claude agents` (v2.1.212+) | +| `/fork [prompt]` | Copy the current conversation into a new **background session** and keep working here; the two are independent from that point on and the copy gets its own row in `claude agents` (v2.1.212+). Except when the copy edits in place, Claude Code instructs it to create a worktree of its own before making code changes (isolation instruction requires v2.1.221+) | | `/subtask ` | Spawn a **forked subagent** that inherits the full conversation and works on the task while you keep going; its result returns to this conversation when it finishes (v2.1.212+) | | `/btw ` | Ask an ephemeral side question while Claude is working on the main task; doesn't pollute the main conversation context | | `/cd ` | Move the session to a new working directory without breaking the prompt cache (added v2.1.169) | @@ -72,13 +72,13 @@ Built-in commands are shortcuts for common actions. There are **60+ built-in com | `/privacy-settings` | Privacy settings (Pro/Max only) | | `/release-notes` | View changelog | | `/recap` | Show session recap / summary when returning to a session (added v2.1.108) | -| `/reload-plugins` | Reload active plugins | +| `/reload-plugins` | Reload active plugins. Since v2.1.221 most installs activate immediately, so this is only needed when the install summary says `Run /reload-plugins to activate.` | | `/reload-skills` | Re-scan skill directories without restarting the session (added v2.1.152) | | `/remote-control` | Remote control from claude.ai (alias: `/rc`) | | `/remote-env` | Configure default remote environment | | `/rename [name]` | Rename session | | `/resume [session]` | Resume conversation (alias: `/continue`) | -| `/review ` | Review a GitHub PR. As of v2.1.186 it runs on the same review engine as `/code-review medium`. Use `/code-review` to review the local working diff | +| `/review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | Alias of `/code-review` (v2.1.223): reviews the current diff, or a PR number, branch, or path you pass — e.g. `/review 1234`. Takes the same effort levels and flags. With no level given, it reuses the last `low`–`max` level you typed | | `/rewind` | Rewind conversation and/or code (alias: `/checkpoint`) | | `/sandbox` | Toggle sandbox mode | | `/schedule [description]` | Create/manage Cloud scheduled tasks | @@ -87,15 +87,15 @@ Built-in commands are shortcuts for common actions. There are **60+ built-in com | `/skills` | List available skills | | `/stats` | Typing-shortcut alias for `/usage` — opens the stats tab (daily usage, sessions, streaks) (v2.1.118+) | | `/stickers` | Order Claude Code stickers | -| `/status` | Show version, model, account | +| `/status` | Show version, model, account, and a `Session kind` row reading `background job · attached`, `background job · unattended`, or `interactive` (row added v2.1.221). Openable while Claude is responding | | `/statusline` | Configure status line | | `/tasks` | List/manage background tasks | | `/team-onboarding` | Generate a teammate ramp-up guide from the project's Claude Code setup (new in v2.1.101) | +| `/teleport` | Resume a Claude Code on the web session in this terminal; opens a picker of your web sessions (alias: `/tp`). Requires a claude.ai subscription | | `/terminal-setup` | Configure terminal keybindings | | `/theme` | Open theme picker / manage custom themes (v2.1.118). Define custom themes via JSON in `~/.claude/themes/.json` | | `/tui` | Toggle fullscreen TUI (text user interface) mode with flicker-free rendering (added v2.1.110) | -| `/ultraplan ` | Draft plan in ultraplan session, review in browser | -| `/ultrareview` | Comprehensive cloud-based code review with multi-agent analysis (added v2.1.111) | +| `/ultrareview` | Comprehensive cloud-based multi-agent code review (added v2.1.111). The preferred invocation is now `/code-review ultra`; `/ultrareview` remains as an alias. Includes 3 free runs on Pro and Max, then requires usage credits | | `/undo` | Alias for `/rewind` (added v2.1.108) | | `/upgrade` | Open upgrade page for higher plan tier | | `/usage` | Canonical usage dashboard (v2.1.118) — combines plan usage limits, rate limits, cost, and daily session stats. `/cost` and `/stats` are typing-shortcut aliases that open specific tabs | @@ -115,7 +115,7 @@ These skills ship with Claude Code and are invoked like slash commands: | `/dataviz` | Chart and dashboard design guidance with a runnable color-palette validator (v2.1.198) | | `/debug [description]` | Enable debug logging | | `/loop [interval] ` | Run prompt repeatedly on interval | -| `/code-review [effort]` | Review the current diff for correctness bugs at a chosen effort level (e.g. `/code-review high`). Originally absorbed `/simplify` in v2.1.146, but `/simplify` returned as a distinct command in v2.1.154 | +| `/code-review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | Review the current diff — or a PR number, branch, or path you pass — for correctness bugs. Pass `--fix` to apply findings, `--comment` to post them as inline GitHub PR comments, or `ultra` to run a deep cloud review; with `ultra` on a `github.com` PR target, `--post` preselects posting the findings to the PR. With no effort level given, the review reuses the last level you typed (v2.1.223). Originally absorbed `/simplify` in v2.1.146, but `/simplify` returned as a distinct command in v2.1.154 | | `/simplify` | Run a cleanup-only review (reuse / simplification / efficiency / altitude) and apply the fixes; does **not** hunt for bugs — use `/code-review` for that. Briefly an alias of `/code-review --fix` (v2.1.152), it became cleanup-only in v2.1.154 | ### Deprecated Commands @@ -131,14 +131,14 @@ These skills ship with Claude Code and are invoked like slash commands: - `/fork` and `/subtask` swapped roles in **v2.1.212**. `/fork` now copies the conversation into a new independent background session; the forked-subagent behavior it used to have moved to the new `/subtask` command. History: `/fork` was an alias for `/branch` from v2.1.77 to v2.1.161; from v2.1.161 to v2.1.211 it started a forked subagent (what `/subtask` does now). When agent view is turned off, `/subtask` is unavailable and `/fork` keeps the forked-subagent behavior - `/resume` (no arguments) opens a picker of past sessions — including ones removed from the visible list — and resumes the chosen one as a background session (v2.1.212) - `/output-style` deprecated (v2.1.73) and removed (v2.1.91) — output styles are still available via `/config` → Output style or the `outputStyle` setting; the built-ins are Default, Proactive, Explanatory, and Learning -- `/review ` now uses the same review engine as `/code-review medium` (v2.1.186) +- `/review` became a full alias of `/code-review` — same targets, effort levels, and flags (v2.1.223). History: it first moved onto the `/code-review medium` engine in v2.1.186 while remaining PR-only - `/effort` command added; `max` level available on Opus 4.6+ (originally Opus 4.6-only) - `/voice` command added for push-to-talk voice dictation - `/schedule` command added for creating/managing scheduled tasks - `/color` command added for prompt bar customization - /pr-comments removed in v2.1.91 — ask Claude directly to view PR comments - /vim removed in v2.1.92 — use /config → Editor mode instead -- /ultraplan added for browser-based plan review and execution +- `/ultraplan` was removed in v2.1.222 — use plan mode instead - /powerup added for interactive feature lessons - /sandbox added for toggling sandbox mode - `/model` picker now shows human-readable labels (e.g., "Sonnet 4.6") instead of raw model IDs @@ -160,6 +160,13 @@ These skills ship with Claude Code and are invoked like slash commands: - `/model` now saves the selected model as the default for new sessions; press `s` for session-only (keybinding `modelPicker:setAsDefault` → `modelPicker:thisSessionOnly`) (v2.1.153) - `/workflows` added — view running and completed dynamic workflow runs (v2.1.154) - `/simplify` returned as a distinct cleanup-only review command (reuse / simplification / efficiency / altitude), separate from `/code-review`'s bug hunt (v2.1.154) +- `/status` gained a `Session kind` row distinguishing attached and unattended background jobs from interactive sessions (v2.1.221) +- Plugin installs now activate immediately when it is safe to do so; `/reload-plugins` is only needed when the install summary asks for it (v2.1.221) +- `/ultraplan` removed — use plan mode (v2.1.222) +- `/code-review` and `/review` remember the last effort level you typed when you omit one (v2.1.223) +- `/code-review ultra` became the preferred entry point for cloud multi-agent review; `/ultrareview` stays as an alias (v2.1.223) +- `/code-review` at `high`, `xhigh`, and `max` effort now runs in a background agent like the other levels (v2.1.232) +- The startup tip suggesting you create custom subagents, and the matching nudge in the `/powerup` tour, were removed (v2.1.232) ### `/goal` — Session-Level Completion Condition @@ -634,10 +641,9 @@ If both exist with the same name, the **skill takes precedence**. Remove one or --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: -- https://code.claude.com/docs/en/slash-commands - https://code.claude.com/docs/en/skills - https://code.claude.com/docs/en/interactive-mode - https://code.claude.com/docs/en/changelog diff --git a/04-subagents/README.md b/04-subagents/README.md index ec0c692..f33d2f4 100644 --- a/04-subagents/README.md +++ b/04-subagents/README.md @@ -579,7 +579,9 @@ graph TB Forked subagents (`context: fork`) inherit the parent agent's full conversation context at the moment of forking, rather than starting with a clean slate. This is useful for exploring alternative paths without losing the work done so far. -> **Availability**: GA in v2.1.117. On external builds (non-first-party distributions), set `CLAUDE_CODE_FORK_SUBAGENT=1` to enable forking. +> **Availability**: GA in v2.1.117. **Since v2.1.232, fork mode is on by default in interactive sessions** — on every build, first-party or not. It stays off by default in non-interactive mode (`claude -p`) and in the Agent SDK. On Claude Code older than v2.1.232, or to turn it on where it is off by default, set `CLAUDE_CODE_FORK_SUBAGENT=1`. + +> **Fork-mode subagents run in the background.** Where fork mode is on — as it is by default in an interactive session — Claude Code runs the subagent in the background, forked and non-forked subagents alike. ### Configuration @@ -596,7 +598,10 @@ may explore an alternative approach. Return your findings and the parent will decide whether to adopt them. ``` -### Enabling on External Builds +### Enabling Fork Mode Explicitly + +Interactive sessions on v2.1.232+ need no flag. Use this on older versions, in headless +runs, or in the Agent SDK: ```bash export CLAUDE_CODE_FORK_SUBAGENT=1 @@ -829,15 +834,11 @@ As of v2.1.210, Claude Code scans each subagent's final report for text that imi When the scan flags something, Claude Code neutralizes it — inserting a backslash or an inline marker such as `[harness: subagent output matched instruction-shaped pattern(s): ...]` naming what triggered the scan — and the parent session is expected to treat the flagged text as a finding to relay, not an instruction to follow. The scan is on by default with no documented opt-out. It errs toward flagging: a legitimate subagent report that quotes a real flag name (e.g. `--dangerously-skip-permissions`) verbatim can trigger a marker even though nothing malicious occurred — a false positive is preferable to a missed injection. -### Session-Wide Subagent Limit (v2.1.212+) +### Subagent Concurrency and Depth Limits -Claude Code caps subagent spawns at **200 per session** by default, to stop runaway delegation loops. Override with `CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION`; the budget resets when you run `/clear`. +> **The per-session spawn cap is gone.** Claude Code capped subagent spawns at 200 per session from v2.1.212, but **v2.1.224 removed that cap** — long-running sessions no longer refuse new agents, and the official subagents reference now states plainly that there is no limit on the total number of subagents Claude can spawn over a session. The `CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION` variable that overrode it is gone with it. -```bash -export CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION=200 -``` - -Two more environment variables cap subagent fan-out: +Two limits on subagent fan-out do still apply, both set by environment variable: - `CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS` (v2.1.217) - Maximum number of subagents running **concurrently** at once. Default: 20. - `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH` (v2.1.217) - Maximum **nesting depth** for subagents spawning their own subagents. **Default: 3 since v2.1.219** (was 1 in v2.1.217–v2.1.218). Set this to `1` to disable nesting (see [Key Behaviors](#key-behaviors)). @@ -1305,8 +1306,8 @@ See the OpenTelemetry section in [Advanced Features → Telemetry](../09-advance --- -**Last Updated**: July 29, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/sub-agents - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md diff --git a/06-hooks/README.md b/06-hooks/README.md index 4a4bcfb..9234f6c 100644 --- a/06-hooks/README.md +++ b/06-hooks/README.md @@ -271,6 +271,13 @@ Claude Code supports **31 hook events**: | **ElicitationResult** | User responds to elicitation | (none) | Yes | Response processing | | **SessionEnd** | Session terminates | (none) | No | Cleanup, final logging | +> **`TaskCreated` and `TaskCompleted` need the todo tools enabled (v2.1.233).** These two +> events fire from the todo/task-tracking tools (`TaskCreate`/`Get`/`Update`/`List`, +> `TodoWrite`), which are **no longer available on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, +> and newer models**. On those models the hooks are still valid configuration but simply +> never fire — you get no output and no error. Set `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` to +> bring the tools, and therefore the events, back. + > **PostToolUse duration (v2.1.119):** `PostToolUse` and `PostToolUseFailure` hook inputs now include `duration_ms` — see the [PostToolUse](#posttooluse) section for details. ### PreToolUse @@ -1517,8 +1524,8 @@ Edit `~/.claude/settings.json` or `.claude/settings.json` with the hook configur --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/hooks - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md diff --git a/07-plugins/README.md b/07-plugins/README.md index a4aa0b7..0a9ac7f 100644 --- a/07-plugins/README.md +++ b/07-plugins/README.md @@ -517,6 +517,17 @@ Enterprise and advanced users can control marketplace behavior through settings: | `blockedMarketplaces` | Admin-managed blocklist of marketplaces (supports `hostPattern` / `pathPattern` regex fields since v2.1.119) | | `deniedPlugins` | Admin-managed blocklist to prevent specific plugins from being installed | +> **Friendlier aliases (v2.1.232)**: `additionalMarketplaces` is accepted as an alias for +> `extraKnownMarketplaces`, and `allowedMarketplaces` for `strictKnownMarketplaces`. +> **Changelog-sourced** — the v2.1.232 changelog announces them, but the official settings +> reference does not yet list either name. The canonical keys are safe to keep using. + +> **Owner wildcards (v2.1.223+)**: a `"owner/*"` entry allows or blocks every marketplace +> repo under one GitHub owner. **Accepted only in `strictKnownMarketplaces` and +> `blockedMarketplaces`.** Everywhere else a `github` source appears — including +> `extraKnownMarketplaces` and `/plugin marketplace add` — the `repo` value must name a +> single repository. + > **Enforcement** (v2.1.117+): `blockedMarketplaces` and `strictKnownMarketplaces` are enforced on every plugin lifecycle event — install, update, refresh, and autoupdate — not just at first add. `strictKnownMarketplaces` is managed-only. Example `blockedMarketplaces` with host/path regex (v2.1.119): @@ -604,9 +615,56 @@ Plugins can be sourced from multiple locations: | **Git subdirectory** | `{ "source": "git-subdir", "url": "...", "path": "..." }` | `{ "source": "git-subdir", "url": "https://github.com/org/monorepo.git", "path": "packages/plugin" }` | | **npm** | `{ "source": "npm", "package": "..." }` | `{ "source": "npm", "package": "@acme/claude-plugin", "version": "^2.0" }` | | **pip** | `{ "source": "pip", "package": "..." }` | `{ "source": "pip", "package": "claude-data-plugin", "version": ">=1.0" }` | +| **Archive** (v2.1.224+) | `{ "source": "archive", "url": "..." }` | `{ "source": "archive", "url": "https://cdn.example.com/lint-plugin-1.2.0.zip", "sha256": "…" }` | +| **Command** (v2.1.229+) | `{ "source": "command", "command": "..." }` | `{ "source": "command", "command": "acme-plugin-resolver --print-dir" }` | GitHub and git sources support optional `ref` (branch/tag) and `sha` (commit hash) fields for version pinning. +#### `archive` source (v2.1.224+) + +Install a plugin from a zip over HTTPS — no git clone, no npm install. + +```json +{ + "source": "archive", + "url": "https://cdn.example.com/lint-plugin-1.2.0.zip", + "sha256": "3b1f0c2e9a7d4f5b8c6e1a2d3f4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b" +} +``` + +| Field | Required | Notes | +|-------|----------|-------| +| `url` | Yes | **HTTPS only.** `http://`, loopback, link-local, and cloud-metadata hosts are rejected — and re-checked on **every redirect hop**, so a redirect cannot smuggle you onto a blocked host | +| `sha256` | No | 64 hex characters. On mismatch the install fails with `Plugin archive integrity check failed` | + +Archives are capped at **256 MiB**. Pin `sha256` for anything you did not build yourself — without it, whoever controls the URL controls the code that runs in your session. + +#### `command` source (v2.1.229+) + +Let a locally installed tool decide where the plugin lives. Useful when an internal +package manager already knows how to fetch and lay out your plugins. + +```json +{ + "source": "command", + "command": "acme-plugin-resolver --print-dir", + "timeout": 60, + "mode": "copy" +} +``` + +The contract is strict: **the command must print exactly one line on stdout and exit with +code 0.** That line is the absolute path of a directory containing the complete plugin. + +| Field | Required | Default | Notes | +|-------|----------|---------|-------| +| `command` | Yes | — | The command to run | +| `timeout` | No | `60` (seconds) | Maximum 600 | +| `mode` | No | `"copy"` | `"copy"` snapshots the directory; `"link"` symlinks it, so edits are live | + +The command is re-resolved each session and the result is applied without a restart. +Organizations can block this source type entirely with `disableCommandPluginSources`. + Reserved marketplace names now include `first-party-plugins` and `healthcare` (v2.1.205) — these are held for official use and cannot be claimed by a custom marketplace. ### Distribution methods @@ -622,6 +680,13 @@ Reserved marketplace names now include `first-party-plugins` and `healthcare` (v /plugin marketplace add https://gitlab.com/org/marketplace-repo.git ``` +Bare `gitlab.com` repo URLs — including nested subgroups — clone the same way `github.com` +URLs do (v2.1.232). **The scheme is mandatory**: since v2.1.196 a bare +`gitlab.example.com/team/plugins` is rejected as an invalid `owner/repo` shorthand, so use +the full `https://gitlab.com/company/plugins.git` form. v2.1.232 also added GitLab +token-family secret redaction and gave the `glab` CLI the same sandbox and credential-path +protection `gh` already had. + **Private repositories**: Supported via git credential helpers or environment tokens. Users must have read access to the repository. **Official marketplace submission**: Submit plugins to the Anthropic-curated marketplace for broader distribution via [claude.ai/settings/plugins/submit](https://claude.ai/settings/plugins/submit) or [platform.claude.com/plugins/submit](https://platform.claude.com/plugins/submit). @@ -745,6 +810,17 @@ LSP servers were added to the details pane in v2.1.142. See also the marketplace claude plugin install plugin-name@marketplace-name ``` +**Does it take effect right away?** Since **v2.1.221**, usually yes — read the last line of +the install summary: + +| Install summary says | What it means | +|---|---| +| `Plugin is now active.` | Claude Code activated the plugin as part of the install. Nothing more to do. | +| `Run /reload-plugins to activate.` | The plugin is installed but not live yet — either activating it would have invalidated the prompt cache, or the activation attempt failed. | + +Before v2.1.221, no install took effect in the current session until you ran +`/reload-plugins` or restarted, so older guides describe that step as unconditional. + ### Enable / Disable (with auto-detected scope) ```bash /plugin enable plugin-name @@ -791,7 +867,7 @@ Claude Code can automatically update marketplaces and their installed plugins at When auto-update runs, Claude Code: 1. Refreshes marketplace catalog 2. Updates installed plugins to latest versions -3. Shows notification prompting `/reload-plugins` +3. Reports the outcome per plugin: `Plugin is now active.` when Claude Code activated it as part of the update, or `Run /reload-plugins to activate.` when it did not ### Environment Variables @@ -889,6 +965,18 @@ Administrators can control plugin behavior across an organization using managed | `strictKnownMarketplaces` | Restrict which marketplaces users are allowed to add (managed-only; enforced on every plugin lifecycle event since v2.1.117) | | `blockedMarketplaces` | Blocklist of marketplaces; enforced on every plugin lifecycle event since v2.1.117; supports `hostPattern` / `pathPattern` regex fields since v2.1.119 | | `allowedChannelPlugins` | Control which plugins are permitted per release channel | +| `disableCommandPluginSources` | Block the `command` plugin source type org-wide (v2.1.229+) | + +> **Friendlier aliases (v2.1.232)**: `additionalMarketplaces` is accepted as an alias for +> `extraKnownMarketplaces`, and `allowedMarketplaces` for `strictKnownMarketplaces`. +> **Changelog-sourced** — the v2.1.232 changelog announces them, but the official settings +> reference does not yet list either name. The canonical keys are safe to keep using. + +> **Owner wildcards (v2.1.223+)**: a `"owner/*"` entry allows or blocks every marketplace +> repo under one GitHub owner. **Accepted only in `strictKnownMarketplaces` and +> `blockedMarketplaces`.** Everywhere else a `github` source appears — including +> `extraKnownMarketplaces` and `/plugin marketplace add` — the `repo` value must name a +> single repository. These settings can be applied at the organization level via managed configuration files and take precedence over user-level settings. @@ -1134,13 +1222,13 @@ The following Claude Code features work together with plugins: --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/plugins - https://code.claude.com/docs/en/changelog#2-1-172 - https://code.claude.com/docs/en/changelog -- https://code.claude.com/docs/en/slash-commands +- https://code.claude.com/docs/en/commands - https://code.claude.com/docs/en/plugin-marketplaces - https://github.com/anthropics/claude-code/releases/tag/v2.1.117 - https://github.com/anthropics/claude-code/releases/tag/v2.1.118 diff --git a/09-advanced-features/README.md b/09-advanced-features/README.md index cbe3edd..720ffe0 100644 --- a/09-advanced-features/README.md +++ b/09-advanced-features/README.md @@ -11,16 +11,16 @@ Comprehensive guide to Claude Code's advanced capabilities including planning mo 1. [Overview](#overview) 2. [Planning Mode](#planning-mode) -3. [Ultraplan (Cloud Plan Drafting)](#ultraplan-cloud-plan-drafting) -4. [Extended Thinking](#extended-thinking) -5. [Auto Mode](#auto-mode) -6. [Background Tasks](#background-tasks) -7. [Monitor Tool (Event-Driven Streams)](#monitor-tool-event-driven-streams) -8. [Dynamic Workflows](#dynamic-workflows) -9. [Scheduled Tasks](#scheduled-tasks) -10. [Permission Modes](#permission-modes) -11. [Headless Mode](#headless-mode) -12. [Session Management](#session-management) +3. [Extended Thinking](#extended-thinking) +4. [Auto Mode](#auto-mode) +5. [Background Tasks](#background-tasks) +6. [Monitor Tool (Event-Driven Streams)](#monitor-tool-event-driven-streams) +7. [Dynamic Workflows](#dynamic-workflows) +8. [Scheduled Tasks](#scheduled-tasks) +9. [Permission Modes](#permission-modes) +10. [Headless Mode](#headless-mode) +11. [Session Management](#session-management) +12. [Cross-Session Messaging](#cross-session-messaging) 13. [Interactive Features](#interactive-features) 14. [Output Styles](#output-styles) 15. [Status Line](#status-line) @@ -37,9 +37,10 @@ Comprehensive guide to Claude Code's advanced capabilities including planning mo 26. [Sandboxing](#sandboxing) 27. [Managed Settings (Enterprise)](#managed-settings-enterprise) 28. [Configuration and Settings](#configuration-and-settings) -29. [Agent Teams](#agent-teams) -30. [Best Practices](#best-practices) -31. [Additional Resources](#additional-resources) +29. [Trust and Permission Scoping](#trust-and-permission-scoping) +30. [Agent Teams](#agent-teams) +31. [Best Practices](#best-practices) +32. [Additional Resources](#additional-resources) --- @@ -209,60 +210,6 @@ claude --model opusplan "design and implement the new API" --- -## Ultraplan (Cloud Plan Drafting) - -> **New in v2.1.101**: Ultraplan now auto-creates a Claude Code on the web cloud environment the first time you invoke it — no manual setup, no waiting for a container to warm up before the draft starts. - -> **Note**: Ultraplan is a research preview and requires Claude Code v2.1.91 or newer. - -`/ultraplan` hands a planning task from your local CLI to a Claude Code on the web session running in plan mode. Claude drafts the plan in the cloud while your terminal stays free for other work, then you review the draft in the browser and choose where to execute — in the same cloud session or teleported back to your terminal. - -### When to Use Ultraplan - -- You want a richer review surface than the terminal: inline comments, emoji reactions, an outline sidebar, and persistent history. -- You want hands-off drafting while you keep coding locally — the cloud session researches the repo and writes the plan without blocking your CLI. -- The plan needs stakeholder review before execution — a shareable web URL beats pasting terminal scrollback. - -### Requirements - -- A Claude Code on the web account. -- A GitHub repository (the cloud session clones your repo to draft against real code). -- **Not available** on Amazon Bedrock, Google Cloud Vertex AI, or Microsoft Foundry. - -### Three Ways to Launch - -- **Command**: `/ultraplan ` — explicit invocation. -- **Keyword**: include the word `ultraplan` anywhere in a normal prompt and Claude routes the request to the cloud. -- **From a local plan**: after Claude finishes a plan locally, pick "No, refine with Ultraplan on Claude Code on the web" in the approval dialog to hand the draft off for deeper research. - -### Usage Example - -```bash -/ultraplan migrate the auth service from sessions to JWTs -``` - -Claude acknowledges, spins up the cloud environment (auto-created on first run in v2.1.101+), and returns a session link you can open in your browser. - -### Status Indicators - -| Status | Meaning | -|---|---| -| `◇ ultraplan` | Claude is researching your codebase and drafting the plan | -| `◇ ultraplan needs your input` | Claude has a clarifying question; open the session link to respond | -| `◆ ultraplan ready` | The plan is ready to review in your browser | - -### Execution Options - -Once the plan is ready, you have two execution paths. Approve the plan in the browser to execute in the same cloud session — Claude implements the changes remotely and opens a pull request from the web UI. Or choose "Approve plan and teleport back to terminal" to implement locally. The terminal teleport dialog offers three choices: - -- **Implement here** — run the approved plan in your current terminal session. -- **Start new session** — open a fresh session in the same working directory and implement there. -- **Cancel** — saves the plan to a file so you can pick it up later. - -> **Warning**: Remote Control disconnects when ultraplan starts. Both features share the claude.ai/code interface, so only one can be active at a time. - ---- - ## Extended Thinking Extended thinking allows Claude to spend more time reasoning about complex problems before providing a solution. @@ -1238,6 +1185,69 @@ CLAUDE_CODE_ENABLE_AWAY_SUMMARY=1 claude # force enable recap --- +## Cross-Session Messaging + +> **Added in v2.1.224**, extended through v2.1.232. Available on **macOS and Linux**. + +Sessions used to be islands. Cross-session messaging lets one Claude Code session talk to +another — including sessions on your other machines and your cloud sessions — so you can +hand a question to the session that already has the right context loaded instead of +re-explaining it. + +### Discovering Sessions + +`ListAgents` lists everything you can address: subagents you spawned, other local sessions +on this machine, your cloud sessions, and (when Remote Control is connected) sessions on +your other machines. Each row is labeled by kind, and since v2.1.229 rows also carry +`offline` and `cloud` labels so you can tell a reachable session from a dormant one. + +The **name in each row is the address** — that is what you send to. + +### Sending a Message + +`SendMessage` takes the target and the message: + +```text +SendMessage({ to: "", message: "What did you conclude about the retry logic?" }) +``` + +Since v2.1.232, a bare name is enough — you no longer need to append a disambiguating ref +unless two rows genuinely share the same name. + +### `@`-Mention Shorthand (v2.1.232) + +Instead of calling the tool explicitly, you can `@`-mention a session directly in your +prompt to reach it: + +```text +@auth-refactor did the migration tests pass? +``` + +### Controlling What Arrives: `crossSessionInbound` + +Inbound messages are governed by the `crossSessionInbound` setting (v2.1.224+): + +| Value | Behavior | +|---|---| +| `"accept"` | Inbound messages are delivered to Claude in this session | +| `"hold"` | You see a notice that a message arrived; it is not delivered | +| `"refuse"` | Inbound messages are dropped | + +The values form a ladder — `accept < hold < refuse` — and **project and local settings +apply only when they are stricter** than the user-scope value. A project can tighten +inbound delivery, never loosen it. Since v2.1.232 the setting also has a `/config` row, +"Messages from your other sessions." + +### Reach and Limits + +- Local sessions on the same machine, plus your cloud sessions. +- Remote Control sessions on your other machines, addressable by name (v2.1.225). +- A cloud session **receives** your message but cannot message a local session back yet — + read its answer in its own transcript. +- macOS and Linux only. + +--- + ## Interactive Features ### Keyboard Shortcuts @@ -1876,15 +1886,24 @@ This starts a Claude Code session on claude.ai that you can access from any brow If you started a session on the web and want to continue it locally: ```bash -# Resume a web session in the local terminal +# Resume a web session in the local terminal — opens a picker of your web sessions claude --teleport ``` Or from within an interactive REPL: -``` + +```text /teleport ``` +`/tp` is an alias for `/teleport`. Both require a claude.ai subscription. Cloud sessions +show a `/teleport` hint explaining how to continue locally (v2.1.223). + +> **Changelog-sourced**: the v2.1.223 changelog shows an argument form, +> `claude --teleport `, that jumps straight to a known session. The CLI +> reference documents only the bare picker form, so prefer `claude --teleport` unless +> you already have a session ID in hand. + ### Use Cases - Start work on one machine and continue on another @@ -2164,6 +2183,43 @@ Example of `deniedDomains` overriding a broad wildcard (v2.1.113+): The wildcard lets everything on `example.com` through, but `deniedDomains` still blocks the specifically-named host. +### Credential Masking (v2.1.221, v2.1.224) + +> **Changelog-sourced**: these `sandbox.credentials` options come from the v2.1.221 and +> v2.1.224 changelog entries; the settings reference does not yet detail them. + +Before v2.1.221, `sandbox.credentials` could only `deny` a credential file — a sandboxed +command that needed the credential simply failed. `mode: "mask"` keeps the command working +without exposing the secret: the sandboxed process reads a **sentinel** copy of the file, +and the sandbox proxy substitutes the real value on the way out to the network. + +```json +{ + "sandbox": { + "network": { "tlsTerminate": true }, + "credentials": { + "files": [ + { "path": "~/.aws/credentials", "mode": "mask" } + ] + } + } +} +``` + +| Capability | Since | What it does | +|---|---|---| +| `mode: "mask"` for credential **files** | v2.1.221 | Sandboxed commands read a sentinel; the proxy swaps in the real value on egress. **Linux and WSL only** — on macOS file masking falls back to `deny`. | +| `extract` / `onExtractNoMatch` | v2.1.224 | Mask one field inside a structured environment value instead of the whole variable, and decide what happens when the pattern doesn't match. | +| `decode: "jwt"` with `maskClaims` | v2.1.224 | Decode a JWT and mask only the named claims, leaving the rest readable. | +| `awsPairs` / `sigv4` | v2.1.224 | Re-sign AWS SigV4 requests at the proxy after substituting the real access key. | + +**Two constraints that are easy to miss:** + +- All masking requires `network.tlsTerminate` — the proxy has to see inside the request to + substitute the value. +- These options are honored **only** from user settings, managed settings, or `--settings`. + Project settings cannot turn masking on or change what gets masked. + ### Example Configuration ```json @@ -2327,6 +2383,8 @@ These keys go in `~/.claude/settings.json` (or a project `.claude/settings.json` |---------|-------------| | `askUserQuestionTimeout` | Auto-continue an unanswered `AskUserQuestion` dialog after an idle interval. As of **v2.1.200** dialogs no longer auto-continue by default — set this to opt back into timed auto-continue. | | `enableArtifact` | Per-user enable/disable of the Artifact tool (v2.1.196). | +| `crossSessionInbound` | (v2.1.224) How inbound [cross-session messages](#cross-session-messaging) are handled — `"accept"`, `"hold"`, or `"refuse"`. Project and local values apply only when *stricter* on the `accept < hold < refuse` ladder. Exposed in `/config` as "Messages from your other sessions" since v2.1.232. | +| `dialogExpiry` | (v2.1.224) How long an unanswered dialog stays open. Default `"5m"`; accepts `"60s"`, `"5m"`, `"10m"`, or `"never"`. Overridden by `CLAUDE_CODE_USER_DIALOG_TIMEOUT_MS`. Exposed in `/config` as "Dialog expiry" since v2.1.232. | ### Fallback Models (`fallbackModel`) @@ -2415,8 +2473,21 @@ export CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION=200 # Cap on WebSearch t # Accessibility (v2.1.208) export CLAUDE_AX_SCREEN_READER=1 # Enable plain-text screen reader rendering mode. Same effect as --ax-screen-reader or "axScreenReader": true in settings. + +# Newer variables (v2.1.221–v2.1.233) — changelog-sourced; the CLI reference has no env-var section +export CLAUDE_CODE_ENABLE_TODO_TOOLS=1 # (v2.1.233) Restore the todo/task-tracking tools (TaskCreate/Get/Update/List, TodoWrite), which are off on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models +export CLAUDE_CODE_WEBFETCH_CACHE_TTL_MS=900000 # (v2.1.233) WebFetch URL cache TTL. Default 15 minutes. +export CLAUDE_CODE_TOOL_MEMORY_LIMIT=2G # (v2.1.233, Linux) Opt-in memory cgroup applied to Bash commands +export ANTHROPIC_BEDROCK_REGION_PREFIX=us # (v2.1.224) Prefer a specific Bedrock cross-region inference profile +export CLAUDE_CODE_DISABLE_UNKNOWN_MODEL_WINDOW_ENFORCEMENT=1 # (v2.1.223) Restore pre-v2.1.223 auto-compact behavior on unrecognized model IDs +export CLAUDE_CODE_WORKFLOW_PREFIX_STAGGER_MS=0 # (v2.1.229) Disable prefix staggering on dynamic-workflow fan-out +export CLAUDE_CODE_USER_DIALOG_TIMEOUT_MS=300000 # (v2.1.224) Overrides the dialogExpiry setting ``` +> **v2.1.223 — `CLAUDE_CODE_DISABLE_1M_CONTEXT` widened**: the variable now holds **every** +> Claude model with a native 1M-token window down to 200K via auto-compaction, rather than +> only a fixed list of model IDs. + > **v2.1.108**: `ENABLE_PROMPT_CACHING_1H=1` — use a 1-hour prompt cache TTL instead of the default 5-minute TTL. Reduces cache misses in long, stable sessions. (v2.1.129 fixes a regression where the 1-hour TTL was silently downgraded to 5 minutes.) > **v2.1.129**: `CLAUDE_CODE_FORCE_SYNC_OUTPUT=1` forces synchronous output for terminals whose capability auto-detection fails (e.g., Emacs `eat`). `CLAUDE_CODE_PACKAGE_MANAGER_AUTO_UPDATE=1` enables background upgrades on Homebrew/WinGet installs, which otherwise never auto-update. @@ -2473,6 +2544,31 @@ Create `.claude/config.json` in your project: --- +## Trust and Permission Scoping + +> **Changelog-sourced (v2.1.222, v2.1.232)**: these tightenings come from the changelog; +> the settings reference does not yet spell them out. + +A recurring theme in recent releases: security-relevant settings can no longer be widened +by a repository you cloned. Three changes to know about. + +**Nested repositories need their own trust confirmation (v2.1.232).** A git repository +inside a trusted parent directory no longer inherits that trust. If you trust +`~/work/monorepo` and it contains a vendored submodule, you will be asked to trust the +submodule separately the first time Claude Code works inside it. + +**`sandbox.ripgrep` is user-scope only (v2.1.232).** The setting that names the ripgrep +binary the sandbox uses is honored only from user settings, managed settings, or +`--settings`. Project settings can no longer point the sandbox at a different binary. + +**Remote Control auto-start is user-scope only (v2.1.222).** Repo-local settings cannot +enable Remote Control auto-start; it can only be turned on at user scope via `/config`. + +The pattern to internalize: if a setting would let a checked-in file expand what Claude +Code is allowed to do on your machine, assume it is now user-scope only. + +--- + ## Agent Teams Agent Teams is an experimental feature that enables multiple Claude Code instances to collaborate on a task. It is disabled by default. @@ -2588,11 +2684,13 @@ For more information about Claude Code and related features: --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/settings - https://code.claude.com/docs/en/sandboxing +- https://code.claude.com/docs/en/commands +- https://code.claude.com/docs/en/cli-reference - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md - https://code.claude.com/docs/en/model-config - https://code.claude.com/docs/en/permission-modes diff --git a/10-cli/README.md b/10-cli/README.md index 5cb137d..cf75e14 100644 --- a/10-cli/README.md +++ b/10-cli/README.md @@ -61,7 +61,8 @@ The older JavaScript bundle is still produced for Windows and for environments t | `claude install [version]` | Install a specific native-binary version. Accepts `stable`, `latest`, or an explicit version string | `claude install 2.1.131` | | `claude project purge [path]` | Delete all local Claude Code state for a project (transcripts, tasks, debug logs, file-edit history, prompt history, and `~/.claude.json` entry). Omit `[path]` for an interactive picker. Flags: `--dry-run` to preview, `-y/--yes` to skip confirmation, `-i/--interactive` to confirm each item, `--all` for every project (v2.1.126+) | `claude project purge ~/work/repo --dry-run` | | `claude plugin prune` | Remove orphaned auto-installed plugin dependencies (parent plugin gone). `plugin uninstall --prune` does the same cascade after uninstalling a target (v2.1.121+) | `claude plugin prune` | -| `claude ultrareview [target]` | Run `/ultrareview` non-interactively. Prints findings to stdout, exits 0 on success / 1 on failure. Use `--json` for raw payload, `--timeout ` to override the 30-minute default (v2.1.120+) | `claude ultrareview 1234 --json` | +| `claude ultrareview [target]` | Run `/ultrareview` non-interactively. Prints findings to stdout, exits 0 on success / 1 on failure. Use `--json` for raw payload, `--timeout ` to override the 30-minute default, and `--post` / `--no-post` to control whether findings are posted back to the PR. **Requires Claude Code v2.1.227 or later** | `claude ultrareview 1234 --json --no-post` | +| `claude self-hosted-runner ` | Turn your own machine or container into a place Claude Code web, mobile, and desktop sessions can run. `setup` provisions the runner, `doctor` diagnoses it, `orchestrator` runs the coordinating process. **Team and Enterprise plans; requires Claude Code v2.1.224 or later.** On Windows, startup requires an explicit `--base-dir` (v2.1.229) | `claude self-hosted-runner setup` | | `claude auth login` | Log in (supports `--email`, `--sso`). Since v2.1.126, accepts the OAuth code pasted into the terminal as a fallback when the browser callback can't reach localhost (WSL2, SSH, containers) | `claude auth login --email user@example.com` | | `claude auth logout` | Log out of current account | `claude auth logout` | | `claude auth status` | Check auth status (exit 0 if logged in, 1 if not) | `claude auth status` | @@ -74,12 +75,12 @@ The older JavaScript bundle is still produced for Windows and for environments t | `-c, --continue` | Load most recent conversation | `claude --continue` | | `-r, --resume` | Resume specific session by ID or name | `claude --resume auth-refactor` | | `-v, --version` | Output version number | `claude -v` | -| `-w, --worktree` | Start in isolated git worktree | `claude -w` | +| `-w, --worktree` | Start in isolated git worktree. Accepts a GitLab merge-request URL as well as a GitHub PR URL since v2.1.233 | `claude -w` | | `-n, --name` | Session display name | `claude -n "auth-refactor"` | | `--from-pr ` | Resume sessions linked to a pull/merge request. Accepts GitHub (cloud + Enterprise), GitLab MR, and Bitbucket PR URLs since v2.1.119; previously GitHub.com only | `claude --from-pr 42` or `claude --from-pr https://gitlab.example.com/org/repo/-/merge_requests/17` | | `--remote "task"` | Create web session on claude.ai | `claude --remote "implement API"` | | `--remote-control, --rc` | Interactive session with Remote Control | `claude --rc` | -| `--teleport` | Resume web session locally | `claude --teleport` | +| `--teleport` | Resume a web session locally. Bare form opens a picker of your web sessions; the v2.1.223 changelog also shows a `claude --teleport ` form, which the CLI reference does not document. Requires a claude.ai subscription | `claude --teleport` | | `--teammate-mode` | Agent team display mode | `claude --teammate-mode tmux` | | `--bare` | Minimal mode (skip hooks, skills, plugins, MCP, auto memory, CLAUDE.md) | `claude --bare` | | `--safe-mode` | Start with all customizations disabled (CLAUDE.md, plugins, skills, hooks, MCP) to isolate config problems; also `CLAUDE_CODE_SAFE_MODE=1` (v2.1.169) | `claude --safe-mode` | @@ -502,6 +503,8 @@ When you dispatch a session from the view (or via `claude --bg `), you c Sessions that finish their work but leave a background shell open move from "Working" to "Completed" (v2.1.141 fix). Within an attached agent session, `Shift+Tab` cycles through permission modes including auto mode (v2.1.143). +**GitLab merge requests (v2.1.233)** — the Agent View recognizes GitLab MR URLs alongside GitHub PR URLs, and displays merge requests as `!N` (GitHub pull requests stay `#N`). The same release taught `--worktree` to accept a GitLab MR URL. + **Pin a session** — press `Ctrl+T` on a session in `claude agents` to pin it (v2.1.147). Pinned background sessions stay alive when idle, are restarted in place to apply Claude Code updates, and are shed under memory pressure only after non-pinned sessions. (This `Ctrl+T` is scoped to the Agent View; in the main session it toggles the task list view.) --- @@ -569,15 +572,15 @@ pipeline { } ``` -**Headless `ultrareview` (v2.1.120+):** +**Headless `ultrareview` (requires v2.1.227+):** ```yaml # .github/workflows/ultrareview.yml - name: Claude ultrareview - run: claude ultrareview ${{ github.event.pull_request.number }} --json > review.json + run: claude ultrareview ${{ github.event.pull_request.number }} --json --no-post > review.json ``` -`claude ultrareview` exits 0 on a clean review and 1 when findings are reported, so it's a drop-in PR gate. Use `--timeout ` to override the 30-minute default. +`claude ultrareview` exits 0 on a clean review and 1 when findings are reported, so it's a drop-in PR gate. Use `--timeout ` to override the 30-minute default. `--post` posts the finished findings to the pull request; `--no-post` keeps them in stdout only, which is what you want when a later CI step formats the report itself. ### 2. Script Piping @@ -841,7 +844,7 @@ The "ultrathink" keyword in prompts activates deep reasoning. The `/effort` menu | `CLAUDE_CODE_PERFORCE_MODE` | Set to `1` to enable Perforce mode — treats files as read-only by default (for Perforce/P4 version control workflows) (added v2.1.98) | | `DISABLE_UPDATES` | Blocks all update paths including manual `claude update`. Stricter than `DISABLE_AUTOUPDATER`, which only blocks the background autoupdater (v2.1.118+) | | `CLAUDE_CODE_HIDE_CWD` | When set to `1`, hides the current working directory in the startup logo (privacy / screen-share use) (v2.1.119+) | -| `CLAUDE_CODE_FORK_SUBAGENT` | Set to `1` to enable forked subagents on external builds (Bedrock, Vertex, Foundry). No effect on Anthropic API where forked subagents are GA (v2.1.117+) | +| `CLAUDE_CODE_FORK_SUBAGENT` | Set to `1` to turn fork mode on where it is off by default: non-interactive mode (`claude -p`), the Agent SDK, or Claude Code older than v2.1.232. Since v2.1.232 fork mode is on by default in interactive sessions on every build, first-party or not (GA v2.1.117) | | `CLAUDE_CODE_DISABLE_ALTERNATE_SCREEN` | Set to `1` to opt out of the fullscreen alternate-screen renderer; the session stays in normal terminal scrollback. Useful when piping transcripts to logs or pairing with `script(1)` (v2.1.132+). | | `CLAUDE_CODE_SESSION_ID` | Set in every Bash tool subprocess launched by Claude Code; equals the `session_id` in hook input JSON. Use to correlate bash logs with hook telemetry (v2.1.132+). | | `CLAUDE_CODE_ENABLE_FEEDBACK_SURVEY_FOR_OTEL` | Set to `1` to re-enable Anthropic's session-quality survey for organizations capturing OpenTelemetry data. Off by default in OTEL deployments (v2.1.136+). | @@ -855,7 +858,6 @@ The "ultrathink" keyword in prompts activates deep reasoning. The `/effort` menu | `CLAUDE_CODE_ENABLE_GATEWAY_MODEL_DISCOVERY` | Set to `1` to opt in to gateway `/v1/models` discovery when `ANTHROPIC_BASE_URL` is set. Without it, `/model` shows the built-in static list (v2.1.129+) | | `CLAUDE_CODE_ENABLE_AUTO_MODE` | Legacy opt-in for auto mode on Bedrock, Vertex, and Foundry (v2.1.158–v2.1.206). As of v2.1.207, auto mode is available by default on those providers for Sonnet 5, Opus 4.7/4.8, and Fable 5 (Opus 5 added in v2.1.219) — this variable is accepted for compatibility but has no effect | | `CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION` | Cap on WebSearch tool calls per session, to stop runaway search loops. Default 200 (v2.1.212) | -| `CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION` | Cap on subagent spawns per session, to stop runaway delegation loops. Default 200; `/clear` resets the budget (v2.1.212) | | `CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS` | Cap on subagents running concurrently. Default 20 (v2.1.217) | | `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH` | Controls how deep nested subagent spawns can go. Since v2.1.219 the default is **3 layers** (was 1); set to `1` to disable nesting entirely | | `CLAUDE_CODE_MCP_AUTO_BACKGROUND_MS` | Threshold, in milliseconds, before a long-running MCP tool call auto-backgrounds. Default 120000 (2 minutes) (v2.1.212) | @@ -866,6 +868,19 @@ The "ultrathink" keyword in prompts activates deep reasoning. The `/effort` menu | `CLAUDE_ENABLE_STREAM_WATCHDOG` | Streaming idle watchdog (aborts/retries after 5 min with no stream events) is on by default for all providers; set to `0` to disable (v2.1.196). | | `CLAUDE_CODE_MCP_TOOL_IDLE_TIMEOUT` | Override the 5-minute idle abort for remote MCP tool calls that hang with no response (v2.1.187+). | | `CLAUDE_CODE_OPUS_4_6_FAST_MODE_OVERRIDE` | **Removed (no-op as of v2.1.160).** Previously pinned Fast Mode (`/fast`) to Opus 4.6. As of v2.1.219, `/fast` applies to **Opus 5 and Opus 4.8** only — Opus 4.6 and Opus 4.7 are no longer fast-mode targets. | +| `CLAUDE_CODE_ENABLE_TODO_TOOLS` | Set to `1` to restore the todo/task-tracking tools (`TaskCreate`/`Get`/`Update`/`List`, `TodoWrite`), which are unavailable on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models (v2.1.233) | +| `CLAUDE_CODE_WEBFETCH_CACHE_TTL_MS` | How long WebFetch caches a fetched URL. Default 15 minutes (v2.1.233) | +| `CLAUDE_CODE_TOOL_MEMORY_LIMIT` | Linux only: opt in to a memory cgroup applied to Bash commands (v2.1.233) | +| `ANTHROPIC_BEDROCK_REGION_PREFIX` | Prefer a specific Bedrock cross-region inference profile (v2.1.224) | +| `CLAUDE_CODE_DISABLE_UNKNOWN_MODEL_WINDOW_ENFORCEMENT` | Set to `1` to restore pre-v2.1.223 auto-compact behavior on unrecognized model IDs (v2.1.223) | +| `CLAUDE_CODE_WORKFLOW_PREFIX_STAGGER_MS` | Set to `0` to disable prefix staggering on dynamic-workflow fan-out (v2.1.229) | +| `CLAUDE_CODE_USER_DIALOG_TIMEOUT_MS` | Overrides the `dialogExpiry` setting (v2.1.224) | + +> **These seven rows are changelog-sourced.** The CLI reference page has no dedicated +> environment-variable section, so they are documented from the v2.1.221–v2.1.233 +> changelog entries rather than a reference page. + +> **`CLAUDE_CODE_DISABLE_1M_CONTEXT` widened in v2.1.223**: it now holds **every** Claude model with a native 1M-token window to 200K via auto-compaction, not just a fixed list of model IDs. > **`ENABLE_TOOL_SEARCH` on Vertex AI (v2.1.119+)**: Tool search is **disabled by default on Google Cloud Vertex AI** deployments. Users who want the tool-search capability on Vertex must explicitly opt in with `export ENABLE_TOOL_SEARCH=true`. On direct Anthropic API it remains enabled by default. @@ -997,8 +1012,8 @@ claude -p --output-format json "query" --- -**Last Updated**: July 29, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/cli-reference - https://code.claude.com/docs/en/env-vars @@ -1008,7 +1023,7 @@ claude -p --output-format json "query" - https://code.claude.com/docs/en/settings - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md - https://code.claude.com/docs/en/troubleshooting -- https://code.claude.com/docs/en/slash-commands +- https://code.claude.com/docs/en/commands - https://code.claude.com/docs/en/model-config - https://platform.claude.com/docs/en/about-claude/models/overview - https://www.anthropic.com/news/claude-opus-4-8 diff --git a/CATALOG.md b/CATALOG.md index ed19ed6..ffa12ff 100644 --- a/CATALOG.md +++ b/CATALOG.md @@ -66,7 +66,7 @@ Commands are user-invoked shortcuts that execute specific actions. | `/logout` | Sign out | Switch accounts | | `/sandbox` | Toggle sandbox mode | Safe command execution | | `/doctor` | Run diagnostics | Troubleshoot issues | -| `/reload-plugins` | Reload installed plugins | Plugin management | +| `/reload-plugins` | Reload installed plugins. Since v2.1.221 most installs activate immediately; only needed when the install summary says `Run /reload-plugins to activate.` | Plugin management | | `/reload-skills` | Re-scan skill directories without restarting (v2.1.152) | Skill management | | `/workflows` | View running and completed dynamic workflow runs (v2.1.154) | Multi-agent orchestration | | `/release-notes` | Show release notes | Check new features | @@ -79,7 +79,7 @@ Commands are user-invoked shortcuts that execute specific actions. | `/tui` | Toggle fullscreen TUI (text user interface) mode | Flicker-free rendering in fullscreen/tmux | | `/tasks` | View background tasks | Monitor async operations | | `/copy` | Copy last response to clipboard | Share output quickly | -| `/teleport` | Transfer session to another machine | Continue work remotely | +| `/teleport` (alias `/tp`) | Resume a Claude Code on the web session in this terminal; opens a picker. Requires a claude.ai subscription | Continue work remotely | | `/desktop` | Open Claude Desktop app | Switch to desktop interface | | `/theme` | Change color theme; v2.1.118 added custom named themes via `~/.claude/themes/.json` (plugins can ship a `themes/` dir) | Customize appearance | | `/usage` | Canonical command for usage/cost/stats — merged `/cost` and `/stats` into a single tabbed view (v2.1.118); as of v2.1.149 the cost view breaks spending down by category (skills, subagents, plugins, per-MCP-server). In the **VSCode extension** (v2.1.174), the `/usage` (Account & usage) dialog adds an attribution breakdown — cache misses, long-context cost, subagents, and per-skill / per-agent / per-plugin / per-MCP usage over 24h and 7d windows | Monitor quota and costs | @@ -94,8 +94,7 @@ Commands are user-invoked shortcuts that execute specific actions. | `/undo` | Alias for `/rewind` (v2.1.108) | Same as `/rewind` | | `/upgrade` | Check for updates | Version management | | `/team-onboarding` | Generate a teammate ramp-up guide from this project's Claude Code usage | Onboarding new teammates (v2.1.101) | -| `/ultraplan` | Hand a planning task to a Claude Code web session in plan mode | Heavy planning offload (Research Preview, v2.1.91+) | -| `/ultrareview` | Run a cloud multi-agent code review over your current changes | Deep pre-merge review across multiple agents (v2.1.112) | +| `/code-review ultra` | Run a cloud multi-agent code review over your current changes. `/ultrareview` remains as an alias; `/code-review ultra` is the preferred invocation. Includes 3 free runs on Pro and Max, then requires usage credits | Deep pre-merge review across multiple agents (v2.1.112) | | `/fewer-permission-prompts` | Scan transcripts and propose a prioritized allowlist for common read-only tools | Reduce repeat permission prompts in a project (v2.1.112) | ### Custom Commands (Examples) @@ -251,7 +250,7 @@ cp -r 03-skills/* ~/.claude/skills/ | `/loop` | Run prompts on interval | Recurring tasks | | `/run` *(v2.1.145+)* | Launch this project's app to see a change running | Verifying a change in the real app | | `/run-skill-generator` *(v2.1.145+)* | Teach `/run`/`/verify` how to handle a specific project | First-time project setup for `/run` | -| `/code-review` | Review the current diff for correctness bugs at a chosen effort level (e.g. `/code-review high`); pass `--comment` to post findings as inline PR comments. Runs as a **background subagent** since v2.1.218, so review output no longer fills the conversation and stacked slash commands remain its review target | After writing code, before landing a PR | +| `/code-review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | Review the current diff — or a PR, branch, or path — for correctness bugs. `--fix` applies findings, `--comment` posts them as inline PR comments. With no level given it reuses the last one you typed (v2.1.223). Lower effort levels moved to a **background subagent** in v2.1.218; `high`, `xhigh`, and `max` followed in v2.1.232, so review output no longer fills the conversation | After writing code, before landing a PR | | `/simplify` *(distinct again since v2.1.154)* | Cleanup-only review (reuse / simplification / efficiency / altitude) that applies the fixes; does not hunt bugs | Tidying code without a bug hunt | | `/verify` *(v2.1.145+)* | Build, run, and observe the app to confirm a fix works | Validating a fix end-to-end | @@ -366,8 +365,8 @@ Event-driven automation that executes shell commands on Claude Code events. | `Stop` | Claude finishes responding | Response complete | Cleanup, reporting | | `StopFailure` | API error ends turn | API error occurs | Error recovery, logging | | `TeammateIdle` | Teammate agent idle | Agent team coordination | Distribute work | -| `TaskCompleted` | Task marked complete | Task done | Post-task processing | -| `TaskCreated` | Task created via TaskCreate | New task created | Task tracking, logging | +| `TaskCompleted` | Task marked complete (only fires when the todo tools are enabled — see [Hooks](06-hooks/README.md#hook-events)) | Task done | Post-task processing | +| `TaskCreated` | Task created via TaskCreate (only fires when the todo tools are enabled — see [Hooks](06-hooks/README.md#hook-events)) | New task created | Task tracking, logging | | `ConfigChange` | Configuration updated | Settings modified | React to config changes | | `CwdChanged` | Working directory changes | Directory changed | Directory-specific setup | | `FileChanged` | Watched file changes | File modified | File monitoring, rebuild | @@ -466,10 +465,15 @@ cp 02-memory/personal-CLAUDE.md ~/.claude/CLAUDE.md | **Status Line** | Render a custom status line from a command that receives session, model, cost, and context JSON on stdin | `/statusline` or the `statusLine` setting. See [Advanced Features](09-advanced-features/#status-line) | | **Community Marketplace** | Third-party plugins that passed Anthropic's automated validation, each pinned to a commit SHA | `/plugin marketplace add anthropics/claude-plugins-community`, then `/plugin install @claude-community` | | **/team-onboarding** | Auto-generate a teammate ramp-up guide from the project's Claude Code setup (v2.1.101) | Run `/team-onboarding` in your project | -| **Ultraplan auto-create** | Cloud environment created automatically on first `/ultraplan` invocation — no manual setup required (v2.1.101) | Use `/ultraplan ` | | **Remote Control** | Control Claude Code sessions remotely via API | Use the remote control API to send prompts and receive responses programmatically | | **Web Sessions** | Run Claude Code in a browser-based environment | Access via `claude web` or through the Anthropic Console | | **Desktop App** | Native desktop application for Claude Code | Use `/desktop` or download from Anthropic website | +| **Cross-Session Messaging** | Sessions can message each other — including your other machines and cloud sessions — discovered via `ListAgents` (v2.1.224+, macOS/Linux) | See [Advanced Features](09-advanced-features/README.md#cross-session-messaging); control inbound with `crossSessionInbound` | +| **Self-Hosted Runner** | Run Claude Code web, mobile, and desktop sessions on your own machines or containers (v2.1.224+, Team/Enterprise) | `claude self-hosted-runner setup`; see [CLI](10-cli/README.md) | +| **`archive` plugin source** | Install a plugin from an HTTPS zip, optionally pinned by `sha256` (v2.1.224+) | See [Plugins](07-plugins/README.md#archive-source-v21224) | +| **`command` plugin source** | A locally installed tool prints the plugin directory path (v2.1.229+) | See [Plugins](07-plugins/README.md#command-source-v21229) | +| **Marketplace owner wildcards** | `"owner/*"` allows or blocks every marketplace repo under a GitHub owner — `strictKnownMarketplaces` and `blockedMarketplaces` only (v2.1.223+) | See [Plugins](07-plugins/README.md#marketplace-configuration) | +| **Sandbox credential masking** | Sandboxed commands read a sentinel while the proxy substitutes the real secret on egress (v2.1.221+, Linux/WSL) | See [Advanced Features](09-advanced-features/README.md#credential-masking-v21221-v21224) | | **Agent Teams** | Coordinate multiple agents working on related tasks | Configure teammate agents that collaborate and share context | | **Task List** | Background task management and monitoring | Use `/tasks` to view and manage background operations | | **Prompt Suggestions** | Context-aware command suggestions | Suggestions appear automatically based on current context | @@ -549,8 +553,8 @@ chmod +x ~/.claude/hooks/*.sh --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/sub-agents - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md @@ -566,4 +570,7 @@ chmod +x ~/.claude/hooks/*.sh - https://code.claude.com/docs/en/cli-reference - https://code.claude.com/docs/en/model-config - https://code.claude.com/docs/en/skills +- https://code.claude.com/docs/en/plugin-marketplaces +- https://code.claude.com/docs/en/discover-plugins +- https://code.claude.com/docs/en/settings **Compatible Models**: Claude Fable 5, Claude Opus 5, Claude Sonnet 5, Claude Sonnet 4.6, Claude Opus 4.8, Claude Haiku 4.5 diff --git a/INDEX.md b/INDEX.md index 924c0cb..49fd663 100644 --- a/INDEX.md +++ b/INDEX.md @@ -227,7 +227,7 @@ Event-driven automation scripts that execute automatically. **Hook Events** (31, in 4 categories) — when it runs: - Tool Hooks: PreToolUse, PostToolUse, PostToolUseFailure, PostToolBatch, PermissionRequest, PermissionDenied - Session Hooks: SessionStart, Setup, SessionEnd, Stop, StopFailure, SubagentStart, SubagentStop -- Task Hooks: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle +- Task Hooks: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle (TaskCompleted/TaskCreated only fire when the todo tools are enabled — off by default on Opus 4.8, Sonnet 5, Fable 5, Mythos 5 and newer) - Lifecycle Hooks: ConfigChange, CwdChanged, DirectoryAdded, FileChanged, PreCompact, PostCompact, WorktreeCreate, WorktreeRemove, Notification, InstructionsLoaded, Elicitation, ElicitationResult --- @@ -888,8 +888,8 @@ Want to add more examples? Follow the structure: --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/overview - https://code.claude.com/docs/en/hooks diff --git a/QUICK_REFERENCE.md b/QUICK_REFERENCE.md index b72ca0e..8093aaf 100644 --- a/QUICK_REFERENCE.md +++ b/QUICK_REFERENCE.md @@ -131,7 +131,7 @@ claude -r "session" # Resume session by name/ID | **Git Worktrees** | Built-in | `/worktree` | | **Auto Memory** | Built-in | Auto-saves to CLAUDE.md | | **Task List** | Built-in | `/task list` | -| **Bundled Skills (10)** | Built-in | `/batch`, `/claude-api`, `/code-review` *(explicit invocation only since v2.1.215 — Claude won't trigger this on its own)*, `/simplify` *(cleanup-only review; distinct from `/code-review` again since v2.1.154)*, `/debug`, `/fewer-permission-prompts`, `/loop`, `/run` *(v2.1.145+)*, `/run-skill-generator` *(v2.1.145+)*, `/verify` *(v2.1.145+; explicit invocation only since v2.1.215 — Claude won't trigger this on its own)* | +| **Bundled Skills (10)** | Built-in | `/batch`, `/claude-api`, `/code-review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` *(explicit invocation only since v2.1.215 — Claude won't trigger this on its own; with no level given it reuses the last one you typed, v2.1.223)*, `/simplify` *(cleanup-only review; distinct from `/code-review` again since v2.1.154)*, `/debug`, `/fewer-permission-prompts`, `/loop`, `/run` *(v2.1.145+)*, `/run-skill-generator` *(v2.1.145+)*, `/verify` *(v2.1.145+; explicit invocation only since v2.1.215 — Claude won't trigger this on its own)* | --- @@ -412,7 +412,7 @@ cp -r 03-skills/code-review-specialist ~/.claude/skills/ | **/verify** *(v2.1.145+)* | Build, run, and observe the app to confirm a fix works (explicit invocation only since v2.1.215 — Claude won't trigger this on its own) | `/verify` | | **/run-skill-generator** *(v2.1.145+)* | Teach `/run`/`/verify` how to handle a specific project | `/run-skill-generator` | | **Subagent Output Scanning** *(v2.1.210+)* | Scans subagent reports for prompt-injection patterns and neutralizes them | On by default, no opt-out | -| **Session-Wide Spawn Caps** *(v2.1.212, extended v2.1.219)* | 200/session limits on WebSearch calls and subagent spawns; concurrent-subagent cap (default 20) added in v2.1.217; since v2.1.219 subagents can spawn nested subagents up to **depth 3 by default** (v2.1.217 had disabled nesting) | `CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION`, `CLAUDE_CODE_MAX_SUBAGENTS_PER_SESSION`, `CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS` (default 20), `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH` (default 3; set 1 to disable); `/clear` resets | +| **WebSearch Cap and Subagent Fan-Out Limits** *(v2.1.212, extended v2.1.219)* | 200 WebSearch calls per session; concurrent-subagent cap (default 20) added in v2.1.217; since v2.1.219 subagents can spawn nested subagents up to **depth 3 by default** (v2.1.217 had disabled nesting). The 200-subagent-per-session spawn cap was **removed in v2.1.224** — there is no longer any limit on total subagents per session | `CLAUDE_CODE_MAX_WEB_SEARCHES_PER_SESSION` (default 200; `/clear` resets), `CLAUDE_CODE_MAX_CONCURRENT_SUBAGENTS` (default 20), `CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH` (default 3; set 1 to disable) | | **Screen Reader Mode** *(v2.1.208)* | Plain-text rendering mode for screen readers | `--ax-screen-reader` flag, `CLAUDE_AX_SCREEN_READER=1`, or `"axScreenReader": true` in settings | --- @@ -516,8 +516,8 @@ Getting started checklist: --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://code.claude.com/docs/en/cli-reference - https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md diff --git a/README.md b/README.md index 7946f1a..d0914c4 100644 --- a/README.md +++ b/README.md @@ -487,7 +487,7 @@ Configure hooks in `~/.claude/settings.json`: **Hook Events** (31, in 4 categories) — when it runs: - **Tool Hooks**: `PreToolUse`, `PostToolUse`, `PostToolUseFailure`, `PostToolBatch`, `PermissionRequest`, `PermissionDenied` - **Session Hooks**: `SessionStart`, `Setup`, `SessionEnd`, `Stop`, `StopFailure`, `SubagentStart`, `SubagentStop` -- **Task Hooks**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle` +- **Task Hooks**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle` — `TaskCompleted` and `TaskCreated` only fire when the todo tools are enabled, which is off by default on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models (`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` restores them) - **Lifecycle Hooks**: `ConfigChange`, `CwdChanged`, `DirectoryAdded`, `FileChanged`, `PreCompact`, `PostCompact`, `WorktreeCreate`, `WorktreeRemove`, `Notification`, `InstructionsLoaded`, `Elicitation`, `ElicitationResult` @@ -873,9 +873,10 @@ MIT License - see [LICENSE](LICENSE). Free to use, modify, and distribute. The o --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: +- https://code.claude.com/docs/en/hooks - https://code.claude.com/docs/en/overview - https://code.claude.com/docs/en/changelog - https://code.claude.com/docs/en/permission-modes diff --git a/claude_concepts_guide.md b/claude_concepts_guide.md index 60dd1bf..4c95039 100644 --- a/claude_concepts_guide.md +++ b/claude_concepts_guide.md @@ -1154,8 +1154,8 @@ Claude Code supports **31 hook events** across five hook types (command, http, m | **Stop** | Claude finishes responding | Summary generation, cleanup tasks | | **StopFailure** | API error ends turn | Error recovery, logging | | **TeammateIdle** | Agent team teammate idle | Work distribution, coordination | -| **TaskCompleted** | Task marked complete | Post-task processing | -| **TaskCreated** | Task created via TaskCreate | Task tracking, logging | +| **TaskCompleted** | Task marked complete. Only fires when the todo tools are enabled — off by default on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` restores them (v2.1.233) | Post-task processing | +| **TaskCreated** | Task created via TaskCreate. Only fires when the todo tools are enabled — off by default on Opus 4.8, Sonnet 5, Fable 5, Mythos 5, and newer models; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` restores them (v2.1.233) | Task tracking, logging | | **ConfigChange** | Config file changes | Validation, propagation | | **CwdChanged** | Working directory changes | Directory-specific setup | | **DirectoryAdded** | New working directory registered mid-session via `/add-dir` or the SDK `register_repo_root` control request | Set up tooling for a newly added directory | @@ -1474,8 +1474,8 @@ Claude Code supports the following models with adaptive reasoning effort: - [Anthropic Cookbook](https://github.com/anthropics/anthropic-cookbook) --- -**Last Updated**: August 4, 2026 -**Claude Code Version**: 2.1.220 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: - https://www.anthropic.com/news/claude-sonnet-5 - https://code.claude.com/docs/en/cli-reference diff --git a/ja/01-slash-commands/README.md b/ja/01-slash-commands/README.md index 503c34d..4db1e6b 100644 --- a/ja/01-slash-commands/README.md +++ b/ja/01-slash-commands/README.md @@ -72,12 +72,12 @@ | `/privacy-settings` | プライバシー設定（Pro／Max のみ） | | `/release-notes` | 変更履歴を表示 | | `/recap` | セッションへ戻ってきた際にセッションのおさらい／要約を表示（v2.1.108 で追加） | -| `/reload-plugins` | アクティブなプラグインをリロード | +| `/reload-plugins` | アクティブなプラグインをリロード。v2.1.221 以降、ほとんどのインストールは即座に有効化されるため、インストール要約に `Run /reload-plugins to activate.` と表示された場合にのみ必要 | | `/remote-control` | claude.ai からのリモート制御（エイリアス: `/rc`） | | `/remote-env` | デフォルトのリモート環境を設定 | | `/rename [name]` | セッションをリネーム | | `/resume [session]` | 会話を再開（エイリアス: `/continue`） | -| `/review` | **非推奨** — 代わりに `code-review` プラグインを導入する | +| `/review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | `/code-review` のエイリアス（v2.1.223）。現在の diff、または渡した PR 番号・ブランチ・パスをレビューする（例: `/review 1234`）。同じ effort レベルとフラグを受け付ける。レベルを省略すると、直前に入力した `low`〜`max` のレベルを再利用する | | `/rewind` | 会話やコードを巻き戻す（エイリアス: `/checkpoint`） | | `/sandbox` | サンドボックスモードを切り替え | | `/schedule [description]` | クラウドのスケジュールタスクを作成／管理 | @@ -92,8 +92,7 @@ | `/terminal-setup` | ターミナルのキーバインドを設定 | | `/theme` | テーマピッカーを開く／カスタムテーマを管理（v2.1.118）。カスタムテーマは `~/.claude/themes/.json` に JSON で定義 | | `/tui` | ちらつきのないフルスクリーン TUI（テキストユーザーインターフェース）モードを切り替え（v2.1.110 で追加） | -| `/ultraplan ` | ultraplan セッションでプランを下書きし、ブラウザでレビュー | -| `/ultrareview` | マルチエージェント分析による包括的なクラウドベースのコードレビュー（v2.1.111 で追加） | +| `/ultrareview` | マルチエージェント分析による包括的なクラウドベースのコードレビュー（v2.1.111 で追加）。現在の推奨呼び出しは `/code-review ultra` で、`/ultrareview` はそのエイリアスとして残っている。Pro と Max では 3 回まで無料、以降は使用クレジットが必要 | | `/undo` | `/rewind` のエイリアス（v2.1.108 で追加） | | `/upgrade` | 上位プランへのアップグレードページを開く | | `/usage` | 正式な使用状況ダッシュボード（v2.1.118）— プラン使用量制限、レート制限、コスト、日次セッション統計を統合。`/cost` と `/stats` は特定のタブを開くタイピング用エイリアス | @@ -115,7 +114,6 @@ | コマンド | ステータス | |---------|----------| -| `/review` | 非推奨 — `code-review` プラグインに置き換え | | `/output-style` | v2.1.91 で削除 — 出力スタイルは `/config` または `outputStyle` 設定から引き続き利用できる | | `/fork` | `/branch` に改名（エイリアスは引き続き有効、v2.1.77） | | `/pr-comments` | v2.1.91 で削除 — Claude に直接 PR コメントを見るよう依頼する | @@ -125,14 +123,14 @@ - `/fork` を `/branch` に改名し、`/fork` はエイリアスとして残置（v2.1.77） - `/output-style` を削除（v2.1.91）。出力スタイル自体は `/config` または `outputStyle` 設定で引き続き利用できる -- `/review` を非推奨化、`code-review` プラグインを推奨 +- `/review` が `/code-review` の完全なエイリアスになった — ターゲット、effort レベル、フラグはすべて同じ（v2.1.223）。履歴: v2.1.186 でまず `/code-review medium` エンジンに移行したが、当時は PR 専用だった - `/effort` コマンドを追加。`max` レベルは Opus 4.7 が必要（当初は Opus 4.6 限定） - プッシュトゥトーク音声入力の `/voice` コマンドを追加 - スケジュールタスクの作成／管理用の `/schedule` コマンドを追加 - プロンプトバーをカスタマイズする `/color` コマンドを追加 - /pr-comments を v2.1.91 で削除 — Claude に直接 PR コメントを見るよう依頼する - /vim を v2.1.92 で削除 — /config → エディタモードを使う -- ブラウザベースのプランレビューと実行を行う /ultraplan を追加 +- `/ultraplan` は v2.1.222 で削除された — 代わりにプランモードを使う - インタラクティブな機能レッスン用に /powerup を追加 - サンドボックスモード切替用に /sandbox を追加 - `/model` ピッカーが、生のモデル ID ではなく可読ラベル（例: "Sonnet 4.6"）を表示するようになった @@ -145,7 +143,7 @@ - `/undo` を `/rewind` のエイリアスとして追加（v2.1.108） - `/proactive` を `/loop` のエイリアスとして追加（v2.1.105） - `/effort` がインタラクティブな矢印キースライダーを獲得し、`high` と `max` の間に新しい `xhigh` レベルを追加。Opus 4.7 プランではデフォルト思考量が `xhigh` に引き上げられた（v2.1.111） -- 包括的なクラウドベースのマルチエージェントコードレビュー用に `/ultrareview` を追加（v2.1.111） +- 包括的なクラウドベースのマルチエージェントコードレビュー用に `/ultrareview` を追加（v2.1.111）。v2.1.223 で `/code-review ultra` が推奨の呼び出しとなり、`/ultrareview` はエイリアスとして残った - Bash／MCP ツール呼び出しを分析し、`.claude/settings.json` の許可リストで権限プロンプトを減らす `/fewer-permission-prompts` を追加（v2.1.111） - Opus 4.7 を利用する Max サブスクライバについて、Auto モードに `--enable-auto-mode` フラグが不要になった（v2.1.112） @@ -605,10 +603,10 @@ Deploy the application to production: --- -**Last Updated**: April 24, 2026 -**Claude Code Version**: 2.1.119 +**Last Updated**: August 15, 2026 +**Claude Code Version**: 2.1.233 **Sources**: -- https://code.claude.com/docs/en/slash-commands +- https://code.claude.com/docs/en/commands - https://code.claude.com/docs/en/interactive-mode - https://code.claude.com/docs/en/changelog - https://github.com/anthropics/claude-code/releases/tag/v2.1.118 diff --git a/ja/04-subagents/README.md b/ja/04-subagents/README.md index 37f6354..ffe3fb7 100644 --- a/ja/04-subagents/README.md +++ b/ja/04-subagents/README.md @@ -584,7 +584,7 @@ graph TB フォークサブエージェント（`context: fork`）は、フォーク時点の親エージェントの会話コンテキスト全体を継承し、クリーンスレートではなくその状態から開始する。これは、これまでの作業を失わずに代替案を探索する場合に有用である。 -> **可用性**：v2.1.117 で GA。外部ビルド（ファーストパーティ以外の配布）では `CLAUDE_CODE_FORK_SUBAGENT=1` を設定するとフォークが有効化される。 +> **可用性**：v2.1.117 で GA。**v2.1.232 以降、フォークモードは対話セッションではデフォルトで有効**であり、これはファーストパーティかどうかを問わずすべてのビルドに当てはまる。非対話モード（`claude -p`）と Agent SDK ではデフォルト無効のままである。デフォルト無効の環境で有効化する場合、または v2.1.232 より古い Claude Code では、`CLAUDE_CODE_FORK_SUBAGENT=1` を設定する。 ### 設定 diff --git a/ja/06-hooks/README.md b/ja/06-hooks/README.md index 9156b07..2e27d1f 100644 --- a/ja/06-hooks/README.md +++ b/ja/06-hooks/README.md @@ -215,6 +215,8 @@ Claude Code は **31 種類のフックイベント** をサポートする。 | **ElicitationResult** | ユーザーが elicitation に応答 | （なし） | 可 | 応答処理 | | **SessionEnd** | セッション終了 | （なし） | 不可 | クリーンアップ、最終ロギング | +> **`TaskCreated` と `TaskCompleted` は todo ツールが有効なときのみ発火する（v2.1.233）。** これらのイベントは todo／タスク追跡ツール（`TaskCreate`/`Get`/`Update`/`List`、`TodoWrite`）から発火するが、それらは **Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降のモデルでは利用できない**。これらのモデルではフックの設定自体は有効なまま、単に発火しない — 出力もエラーも出ない。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` を設定すると復活する。 + > **PostToolUse の duration（v2.1.119）：** `PostToolUse` と `PostToolUseFailure` のフック入力に `duration_ms` が含まれるようになった。詳細は [PostToolUse](#posttooluse) セクションを参照。 ### PreToolUse @@ -1369,8 +1371,8 @@ chmod +x ~/.claude/hooks/*.sh --- -**最終更新：** 2026 年 4 月 24 日 -**Claude Code バージョン：** 2.1.119 +**最終更新：** 2026 年 8 月 15 日 +**Claude Code バージョン：** 2.1.233 **情報源：** - https://code.claude.com/docs/en/hooks - https://code.claude.com/docs/en/changelog diff --git a/ja/07-plugins/README.md b/ja/07-plugins/README.md index b29ada0..b47acb2 100644 --- a/ja/07-plugins/README.md +++ b/ja/07-plugins/README.md @@ -487,6 +487,10 @@ graph TB | `blockedMarketplaces` | 管理者が管理するマーケットプレイスのブロックリスト（v2.1.119 以降は `hostPattern` / `pathPattern` の正規表現フィールドをサポート） | | `deniedPlugins` | 管理者が管理する、特定のプラグインをインストールさせないブロックリスト | +> **より分かりやすいエイリアス（v2.1.232）**：`additionalMarketplaces` は `extraKnownMarketplaces` の、`allowedMarketplaces` は `strictKnownMarketplaces` のエイリアスとして受け付けられる。**changelog 由来** — v2.1.232 の changelog で告知されたが、公式の settings リファレンスにはまだ記載がない。正式なキーをそのまま使い続けて問題ない。 + +> **オーナーのワイルドカード（v2.1.223 以降）**：`"owner/*"` エントリは、ある GitHub オーナー配下のすべてのマーケットプレイスリポジトリを許可またはブロックする。**`strictKnownMarketplaces` と `blockedMarketplaces` でのみ受け付けられる。** `extraKnownMarketplaces` や `/plugin marketplace add` を含め、`github` ソースが登場する他の箇所では、`repo` の値は単一のリポジトリを指定しなければならない。 + > **適用範囲**（v2.1.117 以降）：`blockedMarketplaces` と `strictKnownMarketplaces` は最初の追加時だけでなく、インストール・更新・リフレッシュ・自動更新といったあらゆるプラグインのライフサイクルイベントで適用される。`strictKnownMarketplaces` は管理者専用。 `blockedMarketplaces` でホスト / パス正規表現を使う例（v2.1.119）： @@ -677,6 +681,15 @@ claude plugin tag # Create a release git tag with ver claude plugin install plugin-name@marketplace-name ``` +**すぐに反映されるか？** **v2.1.221** 以降はたいてい反映される。インストール要約の最終行を確認する： + +| インストール要約の表示 | 意味 | +|---|---| +| `Plugin is now active.` | Claude Code がインストールの一環としてプラグインを有効化した。追加の操作は不要。 | +| `Run /reload-plugins to activate.` | インストールは済んだがまだ有効ではない。有効化するとプロンプトキャッシュが無効になるため、または有効化の試行が失敗したため。 | + +v2.1.221 より前は、`/reload-plugins` を実行するか再起動するまで、どのインストールも現在のセッションには反映されなかった。 + ### 有効化 / 無効化（スコープは自動検出） ```bash /plugin enable plugin-name @@ -707,7 +720,7 @@ Claude Code は起動時にマーケットプレイスとそこにインスト 自動更新が走ると、Claude Code は次の処理を行う： 1. マーケットプレイスのカタログを更新 2. インストール済みプラグインを最新バージョンへ更新 -3. `/reload-plugins` を促す通知を表示 +3. プラグインごとに結果を報告する：更新の一環として有効化された場合は `Plugin is now active.`、そうでない場合は `Run /reload-plugins to activate.` ### 環境変数 @@ -793,6 +806,10 @@ claude --plugin-dir ./my-plugin --plugin-dir ./another-plugin | `blockedMarketplaces` | マーケットプレイスのブロックリスト（v2.1.117 以降はあらゆるライフサイクルイベントで適用、v2.1.119 以降は `hostPattern` / `pathPattern` の正規表現フィールドをサポート） | | `allowedChannelPlugins` | リリースチャネルごとに許可するプラグインを制御 | +> **より分かりやすいエイリアス（v2.1.232）**：`additionalMarketplaces` は `extraKnownMarketplaces` の、`allowedMarketplaces` は `strictKnownMarketplaces` のエイリアスとして受け付けられる。**changelog 由来** — v2.1.232 の changelog で告知されたが、公式の settings リファレンスにはまだ記載がない。正式なキーをそのまま使い続けて問題ない。 + +> **オーナーのワイルドカード（v2.1.223 以降）**：`"owner/*"` エントリは、ある GitHub オーナー配下のすべてのマーケットプレイスリポジトリを許可またはブロックする。**`strictKnownMarketplaces` と `blockedMarketplaces` でのみ受け付けられる。** `extraKnownMarketplaces` や `/plugin marketplace add` を含め、`github` ソースが登場する他の箇所では、`repo` の値は単一のリポジトリを指定しなければならない。 + これらの設定は管理対象設定ファイルで組織レベルに適用でき、ユーザーレベルの設定より優先される。 ## プラグインのセキュリティ @@ -1037,9 +1054,11 @@ Complete PR review workflow with security, testing, and documentation checks. --- -**最終更新**: 2026 年 4 月 24 日 -**Claude Code バージョン**: 2.1.119 +**最終更新**: 2026 年 8 月 15 日 +**Claude Code バージョン**: 2.1.233 **出典**: +- https://code.claude.com/docs/en/discover-plugins +- https://code.claude.com/docs/en/settings - https://code.claude.com/docs/en/plugins - https://code.claude.com/docs/en/plugin-marketplaces - https://github.com/anthropics/claude-code/releases/tag/v2.1.117 diff --git a/ja/09-advanced-features/README.md b/ja/09-advanced-features/README.md index 5fa6f8d..69fcc84 100644 --- a/ja/09-advanced-features/README.md +++ b/ja/09-advanced-features/README.md @@ -15,32 +15,31 @@ Claude Code の高度な機能を網羅的に解説するガイド。プラン 1. [概要](#概要) 2. [プランニングモード](#プランニングモード) -3. [Ultraplan（クラウドでのプラン作成）](#ultraplanクラウドでのプラン作成) -4. [拡張思考](#拡張思考) -5. [オートモード](#オートモード) -6. [バックグラウンドタスク](#バックグラウンドタスク) -7. [Monitor ツール（イベント駆動ストリーム）](#monitor-ツールイベント駆動ストリーム) -8. [スケジュールタスク](#スケジュールタスク) -9. [権限モード](#権限モード) -10. [ヘッドレスモード](#ヘッドレスモード) -11. [セッション管理](#セッション管理) -12. [対話機能](#対話機能) -13. [TUI モード（フルスクリーン）](#tui-モードフルスクリーン) -14. [音声入力](#音声入力) -15. [チャンネル](#チャンネル) -16. [Chrome 連携](#chrome-連携) -17. [リモートコントロール](#リモートコントロール) -18. [Web セッション](#web-セッション) -19. [デスクトップアプリ](#デスクトップアプリ) -20. [タスクリスト](#タスクリスト) -21. [プロンプト候補](#プロンプト候補) -22. [Git ワークツリー](#git-ワークツリー) -23. [サンドボックス](#サンドボックス) -24. [マネージド設定（エンタープライズ）](#マネージド設定エンタープライズ) -25. [設定とコンフィグ](#設定とコンフィグ) -26. [エージェントチーム](#エージェントチーム) -27. [ベストプラクティス](#ベストプラクティス) -28. [追加リソース](#追加リソース) +3. [拡張思考](#拡張思考) +4. [オートモード](#オートモード) +5. [バックグラウンドタスク](#バックグラウンドタスク) +6. [Monitor ツール（イベント駆動ストリーム）](#monitor-ツールイベント駆動ストリーム) +7. [スケジュールタスク](#スケジュールタスク) +8. [権限モード](#権限モード) +9. [ヘッドレスモード](#ヘッドレスモード) +10. [セッション管理](#セッション管理) +11. [対話機能](#対話機能) +12. [TUI モード（フルスクリーン）](#tui-モードフルスクリーン) +13. [音声入力](#音声入力) +14. [チャンネル](#チャンネル) +15. [Chrome 連携](#chrome-連携) +16. [リモートコントロール](#リモートコントロール) +17. [Web セッション](#web-セッション) +18. [デスクトップアプリ](#デスクトップアプリ) +19. [タスクリスト](#タスクリスト) +20. [プロンプト候補](#プロンプト候補) +21. [Git ワークツリー](#git-ワークツリー) +22. [サンドボックス](#サンドボックス) +23. [マネージド設定（エンタープライズ）](#マネージド設定エンタープライズ) +24. [設定とコンフィグ](#設定とコンフィグ) +25. [エージェントチーム](#エージェントチーム) +26. [ベストプラクティス](#ベストプラクティス) +27. [追加リソース](#追加リソース) --- @@ -208,60 +207,6 @@ claude --model opusplan "design and implement the new API" --- -## Ultraplan（クラウドでのプラン作成） - -> **v2.1.101 で新登場：** Ultraplan は初回起動時に Claude Code on the web のクラウド環境を自動作成するようになり、手動セットアップやコンテナ起動の待ち時間なしでドラフト作成を始められる。 - -> **注：** Ultraplan は Research Preview であり、Claude Code v2.1.91 以降が必要である。 - -`/ultraplan` は、ローカル CLI からプランニングタスクを Claude Code on the web のセッション（プランモード）に引き渡す。Claude がクラウドで計画を作成している間、ターミナルは別の作業に使え、ブラウザでドラフトをレビューしてから、同じクラウドセッションで実行するか、ターミナルにテレポートして実行するかを選べる。 - -### Ultraplan を使うべきとき - -- ターミナルより豊かなレビュー環境が欲しいとき：インラインコメント、絵文字リアクション、アウトラインのサイドバー、永続的な履歴を活用できる。 -- ローカルでコーディングを続けながら、ハンズオフでドラフトを進めたいとき：クラウドセッションがリポジトリを調査し計画を書き、CLI をブロックしない。 -- 実行前にステークホルダーのレビューが必要なとき：共有可能な Web URL がターミナルのスクロール内容を貼り付けるより便利。 - -### 必要要件 - -- Claude Code on the web アカウント -- GitHub リポジトリ（クラウドセッションが実コードに対してドラフト作成するためにクローンする） -- Amazon Bedrock、Google Cloud Vertex AI、Microsoft Foundry では **利用不可** - -### 起動方法 3 種 - -- **コマンド：** `/ultraplan ` — 明示的に呼び出す。 -- **キーワード：** 通常のプロンプトに `ultraplan` を含めると、Claude がリクエストをクラウドにルーティングする。 -- **ローカル計画から：** Claude がローカルで計画を作成した後、承認ダイアログで「No, refine with Ultraplan on Claude Code on the web」を選ぶと、より深い調査のためにドラフトを引き渡せる。 - -### 利用例 - -```bash -/ultraplan migrate the auth service from sessions to JWTs -``` - -Claude が応答してクラウド環境を起動し（v2.1.101 以降は初回実行時に自動作成）、ブラウザで開けるセッションリンクを返す。 - -### ステータス表示 - -| ステータス | 意味 | -|---|---| -| `◇ ultraplan` | Claude がコードベースを調査し、計画をドラフト中 | -| `◇ ultraplan needs your input` | Claude が確認質問を持っている。セッションリンクを開いて応答する | -| `◆ ultraplan ready` | 計画がブラウザでレビュー可能になった | - -### 実行オプション - -計画が完成すると、実行経路は 2 つある。ブラウザで計画を承認して同じクラウドセッションで実行する場合、Claude がリモートで変更を実装し、Web UI からプルリクエストを開く。あるいは「Approve plan and teleport back to terminal」を選んでローカルで実装する。ターミナルテレポートのダイアログでは 3 つの選択肢が表示される： - -- **Implement here** — 現在のターミナルセッションで承認済み計画を実行する。 -- **Start new session** — 同じ作業ディレクトリで新規セッションを開いて実行する。 -- **Cancel** — 計画をファイルに保存し、後で再開できるようにする。 - -> **警告：** Ultraplan を開始するとリモートコントロールは切断される。両機能とも claude.ai/code インターフェースを共有しているため、同時に有効化できるのは一方のみである。 - ---- - ## 拡張思考 拡張思考により、Claude は複雑な問題について、解を提示する前により多くの時間をかけて推論できる。 @@ -2166,10 +2111,11 @@ Claude Code および関連機能の詳細： --- -**最終更新：** 2026 年 4 月 24 日 -**Claude Code バージョン：** 2.1.119 +**最終更新：** 2026 年 8 月 15 日 +**Claude Code バージョン：** 2.1.233 **情報源：** - https://code.claude.com/docs/en/permission-modes +- https://code.claude.com/docs/en/commands - https://code.claude.com/docs/en/interactive-mode - https://code.claude.com/docs/en/settings - https://www.anthropic.com/news/claude-opus-4-7 diff --git a/ja/10-cli/README.md b/ja/10-cli/README.md index d3b82cf..8a6de37 100644 --- a/ja/10-cli/README.md +++ b/ja/10-cli/README.md @@ -763,7 +763,7 @@ export CLAUDE_CODE_EFFORT_LEVEL=xhigh # low, medium, high, xhigh (default on O | `CLAUDE_CODE_PERFORCE_MODE` | `1` を設定すると Perforce モードを有効化 — ファイルをデフォルトで読み取り専用として扱う（Perforce/P4 バージョン管理ワークフロー向け）（v2.1.98 追加） | | `DISABLE_UPDATES` | 手動の `claude update` を含むすべての更新経路をブロックする。バックグラウンドの自動更新のみブロックする `DISABLE_AUTOUPDATER` より厳密（v2.1.118+） | | `CLAUDE_CODE_HIDE_CWD` | `1` に設定すると、起動時ロゴでカレント作業ディレクトリを隠す（プライバシー／画面共有用途）（v2.1.119+） | -| `CLAUDE_CODE_FORK_SUBAGENT` | `1` を設定すると外部ビルド（Bedrock、Vertex、Foundry）でフォーク済みサブエージェントを有効化。Anthropic API ではフォーク済みサブエージェントが GA のため効果なし（v2.1.117+） | +| `CLAUDE_CODE_FORK_SUBAGENT` | デフォルトでフォークモードが無効な場面 — 非対話モード（`claude -p`）、Agent SDK、v2.1.232 より古い Claude Code — で `1` を設定するとフォークモードを有効化する。v2.1.232 以降、フォークモードはファーストパーティかどうかを問わずすべてのビルドの対話セッションでデフォルト有効（GA は v2.1.117） | | `OTEL_LOG_TOOL_DETAILS` | `1` を設定すると OpenTelemetry イベント中のカスタムおよび MCP コマンド名のリダクションを解除する（v2.1.117+）。デフォルトはリダクション有効。 | > **Vertex AI 上の `ENABLE_TOOL_SEARCH`（v2.1.119+）**: ツール検索は **Google Cloud Vertex AI 上のデプロイではデフォルト無効** である。Vertex 上でツール検索機能を使いたい場合は `export ENABLE_TOOL_SEARCH=true` で明示的にオプトインする必要がある。Anthropic API 直結ではデフォルト有効のままである。 diff --git a/ja/CATALOG.md b/ja/CATALOG.md index af27645..82fa4ad 100644 --- a/ja/CATALOG.md +++ b/ja/CATALOG.md @@ -70,7 +70,7 @@ | `/logout` | サインアウト | アカウント切替 | | `/sandbox` | サンドボックスモード切替 | 安全なコマンド実行 | | `/doctor` | 診断を実行 | 問題のトラブルシューティング | -| `/reload-plugins` | インストール済みプラグインを再読込 | プラグイン管理 | +| `/reload-plugins` | インストール済みプラグインを再読込。v2.1.221 以降、ほとんどのインストールは即座に有効化されるため、インストール要約が `Run /reload-plugins to activate.` と示した場合にのみ必要 | プラグイン管理 | | `/release-notes` | リリースノートを表示 | 新機能の確認 | | `/remote-control` | リモートコントロールを有効化 | リモートアクセス | | `/permissions` | 権限を管理 | アクセス制御 | @@ -95,8 +95,7 @@ | `/undo` | `/rewind` のエイリアス（v2.1.108） | `/rewind` と同じ | | `/upgrade` | アップデート確認 | バージョン管理 | | `/team-onboarding` | このプロジェクトの Claude Code 利用状況からオンボーディングガイドを生成 | 新メンバーのオンボーディング（v2.1.101） | -| `/ultraplan` | プランニングタスクを Claude Code Web セッションに渡す（plan モード） | 重い計画作業のオフロード（Research Preview、v2.1.91+） | -| `/ultrareview` | 現在の変更に対するクラウドのマルチエージェントコードレビューを実行 | マージ前の深いマルチエージェントレビュー（v2.1.112） | +| `/code-review ultra` | 現在の変更に対するクラウドのマルチエージェントコードレビューを実行。`/ultrareview` はエイリアスとして残っており、`/code-review ultra` が推奨の呼び出し。Pro と Max では 3 回まで無料、以降は使用クレジットが必要 | マージ前の深いマルチエージェントレビュー（v2.1.112） | | `/fewer-permission-prompts` | トランスクリプトをスキャンし、よく使う読取専用ツールの優先許可リストを提案 | プロジェクトでの権限プロンプト繰り返しを削減（v2.1.112） | ### カスタムコマンド（例） @@ -356,8 +355,8 @@ Claude Code のイベントでシェルコマンドを自動実行するイベ | `Stop` | Claude が応答完了 | 応答完了時 | 後処理、レポート | | `StopFailure` | API エラーでターン終了 | API エラー発生時 | エラー復旧、ログ | | `TeammateIdle` | チームメンバーエージェントがアイドル | エージェントチーム協調時 | 作業の分配 | -| `TaskCompleted` | タスク完了マーク | タスク完了時 | タスク後処理 | -| `TaskCreated` | TaskCreate でタスク作成 | 新タスク作成時 | タスク追跡、ログ | +| `TaskCompleted` | タスク完了マーク（todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） | タスク完了時 | タスク後処理 | +| `TaskCreated` | TaskCreate でタスク作成（todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） | 新タスク作成時 | タスク追跡、ログ | | `ConfigChange` | 設定更新 | 設定変更時 | 設定変更への対応 | | `CwdChanged` | 作業ディレクトリ変更 | ディレクトリ変更時 | ディレクトリ別セットアップ | | `FileChanged` | 監視ファイルの変更 | ファイル変更時 | ファイル監視、再ビルド | @@ -453,7 +452,6 @@ cp 02-memory/personal-CLAUDE.md ~/.claude/CLAUDE.md | **/undo** | `/rewind` のエイリアス — 直前のチェックポイントへ戻す（v2.1.108） | `/undo` を `/rewind` と互換に利用 | | **Monitor ツール** | バックグラウンドコマンドの stdout を監視し、ポーリングではなくイベントで反応（v2.1.98+） | [高度な機能](09-advanced-features/) 経由で Monitor ツールを利用 | | **/team-onboarding** | プロジェクトの Claude Code 設定からオンボーディングガイドを自動生成（v2.1.101） | プロジェクトで `/team-onboarding` を実行 | -| **Ultraplan 自動作成** | 初回 `/ultraplan` 実行時にクラウド環境を自動作成、手動セットアップ不要（v2.1.101） | `/ultraplan ` を使う | | **リモートコントロール** | API 経由で Claude Code セッションを遠隔制御 | リモートコントロール API でプロンプト送信と応答取得をプログラム的に行う | | **Web セッション** | ブラウザベース環境で Claude Code を実行 | `claude web` または Anthropic Console でアクセス | | **デスクトップアプリ** | Claude Code のネイティブデスクトップアプリ | `/desktop` または Anthropic サイトからダウンロード | @@ -534,8 +532,8 @@ chmod +x ~/.claude/hooks/*.sh --- -**最終更新**：2026 年 4 月 24 日 -**Claude Code バージョン**：2.1.119 +**最終更新**：2026 年 8 月 15 日 +**Claude Code バージョン**：2.1.233 **情報源**： - https://code.claude.com/docs/en/overview - https://code.claude.com/docs/en/commands diff --git a/ja/INDEX.md b/ja/INDEX.md index a59077b..8287726 100644 --- a/ja/INDEX.md +++ b/ja/INDEX.md @@ -229,7 +229,7 @@ blog-draft/ **フックイベント**（31 件、4 カテゴリ）— 実行タイミング： - ツール系フック：PreToolUse、PostToolUse、PostToolUseFailure、PostToolBatch、PermissionRequest、PermissionDenied - セッション系フック：SessionStart、Setup、SessionEnd、Stop、StopFailure、SubagentStart、SubagentStop -- タスク系フック：UserPromptSubmit、UserPromptExpansion、MessageDisplay、TaskCompleted、TaskCreated、TeammateIdle +- タスク系フック：UserPromptSubmit、UserPromptExpansion、MessageDisplay、TaskCompleted、TaskCreated、TeammateIdle（TaskCompleted と TaskCreated は todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） - ライフサイクル系フック：ConfigChange、CwdChanged、DirectoryAdded、FileChanged、PreCompact、PostCompact、WorktreeCreate、WorktreeRemove、Notification、InstructionsLoaded、Elicitation、ElicitationResult --- @@ -882,16 +882,16 @@ Run tests in background --- -**最終更新**：2026 年 4 月 24 日 -**Claude Code バージョン**：2.1.119 -**情報源**： -- https://code.claude.com/docs/en/overview -- https://code.claude.com/docs/en/hooks -- https://code.claude.com/docs/en/commands -- https://github.com/anthropics/claude-code/releases/tag/v2.1.119 +**最終更新**：2026 年 8 月 15 日 +**Claude Code バージョン**：2.1.233 **互換モデル**：Claude Sonnet 4.6、Claude Opus 4.7、Claude Haiku 4.5 **例題総数**：100 ファイル超 **カテゴリ**：10 機能 **フック**：自動化スクリプト 8 個 **設定例**：10 シナリオ以上 **すぐ使える**：すべての例 +**情報源**： +- https://code.claude.com/docs/en/overview +- https://code.claude.com/docs/en/hooks +- https://code.claude.com/docs/en/commands +- https://github.com/anthropics/claude-code/releases/tag/v2.1.119 diff --git a/ja/README.md b/ja/README.md index ad5d8dd..30e7b05 100644 --- a/ja/README.md +++ b/ja/README.md @@ -491,7 +491,7 @@ chmod +x ~/.claude/hooks/*.sh **フックイベント**（31 件、4 カテゴリ）— 実行タイミング： - **ツール系フック**：`PreToolUse`、`PostToolUse`、`PostToolUseFailure`、`PostToolBatch`、`PermissionRequest`、`PermissionDenied` - **セッション系フック**：`SessionStart`、`Setup`、`SessionEnd`、`Stop`、`StopFailure`、`SubagentStart`、`SubagentStop` -- **タスク系フック**：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle` +- **タスク系フック**：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle`（`TaskCompleted` と `TaskCreated` は todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） - **ライフサイクル系フック**：`ConfigChange`、`CwdChanged`、`DirectoryAdded`、`FileChanged`、`PreCompact`、`PostCompact`、`WorktreeCreate`、`WorktreeRemove`、`Notification`、`InstructionsLoaded`、`Elicitation`、`ElicitationResult` @@ -877,9 +877,10 @@ MIT License — [LICENSE](../LICENSE) を参照。利用、変更、再配布は --- -**最終更新**：2026 年 4 月 24 日 -**Claude Code バージョン**：2.1.119 +**最終更新**：2026 年 8 月 15 日 +**Claude Code バージョン**：2.1.233 **情報源**： +- https://code.claude.com/docs/en/hooks - https://code.claude.com/docs/en/overview - https://code.claude.com/docs/en/changelog - https://github.com/anthropics/claude-code/releases/tag/v2.1.119 diff --git a/ja/claude_concepts_guide.md b/ja/claude_concepts_guide.md index 10f4007..c021377 100644 --- a/ja/claude_concepts_guide.md +++ b/ja/claude_concepts_guide.md @@ -2830,8 +2830,8 @@ Claude Code は 5 種類（command、http、mcp_tool、prompt、agent）にわ | **Stop** | Claude の応答完了時 | サマリー生成、後処理 | | **StopFailure** | API エラーでターン終了時 | エラー回復、ログ | | **TeammateIdle** | エージェントチームのメンバーがアイドル時 | 作業分配、調整 | -| **TaskCompleted** | タスク完了時 | 後続処理 | -| **TaskCreated** | TaskCreate でタスク作成時 | タスク追跡、ログ | +| **TaskCompleted** | タスク完了時（todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） | 後続処理 | +| **TaskCreated** | TaskCreate でタスク作成時（todo ツールが有効なときのみ発火 — Opus 4.8、Sonnet 5、Fable 5、Mythos 5 以降ではデフォルト無効。`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` で復活） | タスク追跡、ログ | | **ConfigChange** | 設定ファイル変更時 | 検証、伝播 | | **CwdChanged** | 作業ディレクトリ変更時 | ディレクトリ別セットアップ | | **FileChanged** | 監視対象ファイルの変更時 | ファイル監視、再ビルド | @@ -3145,13 +3145,13 @@ Claude Code は推論努力を調整可能な 3 モデルをサポートする --- -*最終更新：2026 年 4 月 24 日* +*最終更新：2026 年 8 月 15 日* *対応：Claude Haiku 4.5、Sonnet 4.6、Opus 4.7* *収録機能：フック、チェックポイント、プラニングモード、拡張思考、バックグラウンドタスク、権限モード（6 種）、ヘッドレスモード、セッション管理、Auto Memory、Agent Teams、定期タスク、Chrome 連携、Channels、音声入力、同梱スキル* --- -**最終更新：** 2026 年 4 月 24 日 -**Claude Code バージョン：** 2.1.119 +**最終更新：** 2026 年 8 月 15 日 +**Claude Code バージョン：** 2.1.233 **出典：** - https://code.claude.com/docs/en/overview - https://code.claude.com/docs/en/hooks diff --git a/uk/01-slash-commands/README.md b/uk/01-slash-commands/README.md index 261f066..9fe8579 100644 --- a/uk/01-slash-commands/README.md +++ b/uk/01-slash-commands/README.md @@ -68,12 +68,12 @@ | `/powerup` | Інтерактивні уроки з анімованими демо | | `/privacy-settings` | Налаштування приватності (Pro/Max) | | `/release-notes` | Переглянути журнал змін | -| `/reload-plugins` | Перезавантажити активні плагіни | +| `/reload-plugins` | Перезавантажити активні плагіни. Починаючи з v2.1.221 більшість установок активуються одразу, тож команда потрібна лише коли підсумок установки каже `Run /reload-plugins to activate.` | | `/remote-control` | Віддалене керування з claude.ai (аліас: `/rc`) | | `/remote-env` | Налаштування стандартного віддаленого середовища | | `/rename [name]` | Перейменувати сесію | | `/resume [session]` | Відновити розмову (аліас: `/continue`) | -| `/review` | **Застаріла** — встановіть плагін `code-review` | +| `/review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | Псевдонім `/code-review` (v2.1.223): переглядає поточний diff або номер PR, гілку чи шлях, які ви передали — наприклад `/review 1234`. Приймає ті самі рівні зусиль і прапорці. Якщо рівень не вказано, використовує останній рівень `low`–`max`, який ви вводили | | `/rewind` | Відкат розмови та/або коду (аліас: `/checkpoint`) | | `/sandbox` | Перемкнути режим пісочниці | | `/schedule [description]` | Створити/управляти хмарними запланованими завданнями | @@ -86,7 +86,6 @@ | `/tasks` | Список/управління фоновими завданнями | | `/terminal-setup` | Налаштування клавіш терміналу | | `/theme` | Змінити колірну тему | -| `/ultraplan ` | Створити план в ultraplan-сесії, переглянути в браузері | | `/upgrade` | Відкрити сторінку оновлення тарифу | | `/usage` | Показати ліміти плану та статус обмежень | | `/voice` | Перемкнути голосовий ввід push-to-talk | @@ -107,7 +106,6 @@ | Команда | Статус | |---------|--------| -| `/review` | Застаріла — замінена плагіном `code-review` | | `/output-style` | Видалена в v2.1.91 (застаріла з v2.1.73) — використовуйте `/config` → Output style або налаштування `outputStyle` | | `/fork` | Перейменована на `/branch` (аліас працює, v2.1.77) | | `/pr-comments` | Видалена в v2.1.91 — запитайте Claude напряму | @@ -117,14 +115,14 @@ - `/fork` перейменовано на `/branch`, `/fork` залишено як аліас (v2.1.77) - `/output-style` застаріла (v2.1.73) і видалена (v2.1.91) — стилі виводу все ще доступні через `/config` → Output style або налаштування `outputStyle`; вбудовані: Default, Proactive, Explanatory, Learning -- `/review` застаріла на користь плагіна `code-review` +- `/review` став повним псевдонімом `/code-review` — ті самі цілі, рівні зусиль і прапорці (v2.1.223). Історія: спершу він перейшов на рушій `/code-review medium` у v2.1.186, залишаючись доступним лише для PR - Додано команду `/effort` з рівнем `max` для Opus 4.6 - Додано команду `/voice` для голосового вводу push-to-talk - Додано команду `/schedule` для запланованих завдань - Додано команду `/color` для кастомізації рядка промпту - `/pr-comments` видалена в v2.1.91 - `/vim` видалена в v2.1.92 -- Додано `/ultraplan` для перегляду плану в браузері +- `/ultraplan` вилучено у v2.1.222 — використовуйте режим планування - Додано `/powerup` для інтерактивних уроків - Додано `/sandbox` для режиму пісочниці - Вибір `/model` тепер показує зрозумілі назви (наприклад, "Sonnet 4.6") замість ID моделей @@ -578,8 +576,10 @@ Deploy the application to production: --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/commands **Сумісні моделі**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 *Частина серії посібників [Claude How To](../)* diff --git a/uk/06-hooks/README.md b/uk/06-hooks/README.md index 44d7d2f..6ccb633 100644 --- a/uk/06-hooks/README.md +++ b/uk/06-hooks/README.md @@ -213,6 +213,8 @@ Claude Code підтримує **31 подія хуків**: | **ElicitationResult** | Відповідь користувача на elicitation | (немає) | Так | Обробка відповіді | | **SessionEnd** | Завершення сесії | (немає) | Ні | Очищення, фінальне логування | +> **`TaskCreated` і `TaskCompleted` потребують увімкнених todo-інструментів (v2.1.233).** Ці події виникають від todo / task-tracking інструментів (`TaskCreate`/`Get`/`Update`/`List`, `TodoWrite`), які **більше не доступні на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 та новіших моделях**. На цих моделях хук лишається валідною конфігурацією, але просто ніколи не спрацьовує — без виводу і без помилки. Встановіть `CLAUDE_CODE_ENABLE_TODO_TOOLS=1`, щоб повернути їх. + ### PreToolUse Запускається після створення параметрів інструмента Claude і перед обробкою. Використовується для валідації або модифікації вхідних даних інструмента. @@ -1203,6 +1205,8 @@ chmod +x ~/.claude/hooks/*.sh - **[Посібник з памʼяті](../02-memory/)** — конфігурація постійного контексту --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/hooks **Сумісні моделі**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/uk/07-plugins/README.md b/uk/07-plugins/README.md index 9bd7644..0f1b9d2 100644 --- a/uk/07-plugins/README.md +++ b/uk/07-plugins/README.md @@ -462,6 +462,10 @@ graph TB | `strictKnownMarketplaces` | Контролювати, які маркетплейси дозволено додавати користувачам | | `deniedPlugins` | Блок-список для запобігання встановленню конкретних плагінів (керований адміністратором) | +> **Дружніші аліаси (v2.1.232)**: `additionalMarketplaces` приймається як аліас для `extraKnownMarketplaces`, а `allowedMarketplaces` — для `strictKnownMarketplaces`. **Джерело — changelog**: changelog v2.1.232 оголошує їх, але офіційна довідка з налаштувань поки не містить жодної з цих назв. Канонічні ключі безпечно використовувати й далі. + +> **Шаблони власника (v2.1.223+)**: запис `"owner/*"` дозволяє або блокує всі репозиторії маркетплейсів під одним власником GitHub. **Приймається лише у `strictKnownMarketplaces` та `blockedMarketplaces`.** Скрізь інде, де зʼявляється джерело `github` — включно з `extraKnownMarketplaces` і `/plugin marketplace add` — значення `repo` має вказувати на один репозиторій. + ### Додаткові функції маркетплейсу - **Стандартний таймаут git**: Збільшено з 30с до 120с для великих репозиторіїв плагінів @@ -706,6 +710,10 @@ claude --plugin-dir ./my-plugin --plugin-dir ./another-plugin | `strictKnownMarketplaces` | Обмеження маркетплейсів, які дозволено додавати користувачам | | `allowedChannelPlugins` | Контроль дозволених плагінів для кожного каналу випуску | +> **Дружніші аліаси (v2.1.232)**: `additionalMarketplaces` приймається як аліас для `extraKnownMarketplaces`, а `allowedMarketplaces` — для `strictKnownMarketplaces`. **Джерело — changelog**: changelog v2.1.232 оголошує їх, але офіційна довідка з налаштувань поки не містить жодної з цих назв. Канонічні ключі безпечно використовувати й далі. + +> **Шаблони власника (v2.1.223+)**: запис `"owner/*"` дозволяє або блокує всі репозиторії маркетплейсів під одним власником GitHub. **Приймається лише у `strictKnownMarketplaces` та `blockedMarketplaces`.** Скрізь інде, де зʼявляється джерело `github` — включно з `extraKnownMarketplaces` і `/plugin marketplace add` — значення `repo` має вказувати на один репозиторій. + Ці налаштування можна застосувати на рівні організації через файли керованої конфігурації, і вони мають пріоритет над налаштуваннями рівня користувача. ## Безпека плагінів @@ -833,6 +841,15 @@ Complete PR review workflow with security, testing, and documentation checks. /plugin install plugin-name ``` +**Чи набуде чинності одразу?** Починаючи з **v2.1.221** — зазвичай так. Прочитайте останній рядок підсумку встановлення: + +| Підсумок встановлення каже | Що це означає | +|---|---| +| `Plugin is now active.` | Claude Code активував плагін під час встановлення. Більше нічого робити не треба. | +| `Run /reload-plugins to activate.` | Плагін встановлено, але він ще не активний — або активація знецінила б prompt cache, або спроба активації не вдалася. | + +До v2.1.221 жодне встановлення не набувало чинності в поточній сесії, доки ви не запускали `/reload-plugins` або не перезапускали Claude Code. + ### Встановлення з локального шляху ```bash @@ -948,6 +965,9 @@ Complete PR review workflow with security, testing, and documentation checks. - [Довідник системи хуків](../06-hooks/README.md) --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/discover-plugins +- https://code.claude.com/docs/en/settings **Сумісні моделі**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/uk/09-advanced-features/README.md b/uk/09-advanced-features/README.md index b1dbe16..e45626f 100644 --- a/uk/09-advanced-features/README.md +++ b/uk/09-advanced-features/README.md @@ -197,10 +197,6 @@ claude --model opusplan "design and implement the new API" **Редагування плану зовні**: Натисніть `Ctrl+G`, щоб відкрити поточний план у зовнішньому редакторі для детальних змін. -### Ultraplan - -Використовуйте `/ultraplan ` для наскрізного робочого процесу планування: Claude складає детальний план, відкриває його у браузері для перегляду, потім виконує план віддалено або повертає його у ваш термінал для локального виконання. - --- ## Розширене мислення @@ -1940,6 +1936,8 @@ claude --teammate-mode in-process - [Офіційна документація Agent Teams](https://code.claude.com/docs/en/agent-teams) --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/commands **Сумісні моделі**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/uk/CATALOG.md b/uk/CATALOG.md index 3379891..a6cec6d 100644 --- a/uk/CATALOG.md +++ b/uk/CATALOG.md @@ -68,7 +68,7 @@ | `/logout` | Вийти з облікового запису | Зміна акаунту | | `/sandbox` | Перемкнути режим пісочниці | Безпечне виконання команд | | `/doctor` | Запустити діагностику | Усунення неполадок | -| `/reload-plugins` | Перезавантажити плагіни | Управління плагінами | +| `/reload-plugins` | Перезавантажити плагіни. Починаючи з v2.1.221 більшість установок активуються одразу; команда потрібна лише коли підсумок установки каже `Run /reload-plugins to activate.` | Управління плагінами | | `/release-notes` | Показати нотатки до релізу | Перевірка нових функцій | | `/remote-control` | Увімкнути віддалене керування | Віддалений доступ | | `/permissions` | Управління дозволами | Контроль доступу | @@ -352,8 +352,8 @@ export GITHUB_TOKEN="your_token" && claude mcp add github -- npx -y @modelcontex | `Stop` | Claude завершив відповідь | Відповідь готова | Очищення, звітність | | `StopFailure` | API-помилка завершила черга | API-помилка | Відновлення, логування | | `TeammateIdle` | Агент-партнер бездіє | Координація агентів | Розподіл роботи | -| `TaskCompleted` | Завдання позначено як виконане | Завдання завершено | Пост-обробка | -| `TaskCreated` | Завдання створено через TaskCreate | Нове завдання створено | Відстеження, логування | +| `TaskCompleted` | Завдання позначено як виконане (спрацьовує лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) | Завдання завершено | Пост-обробка | +| `TaskCreated` | Завдання створено через TaskCreate (спрацьовує лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) | Нове завдання створено | Відстеження, логування | | `ConfigChange` | Конфігурацію оновлено | Налаштування змінені | Реакція на зміни конфігурації | | `CwdChanged` | Робочий каталог змінився | Зміна каталогу | Налаштування для каталогу | | `FileChanged` | Відстежуваний файл змінився | Файл модифіковано | Моніторинг, перебудова | @@ -524,5 +524,8 @@ chmod +x ~/.claude/hooks/*.sh --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/commands +- https://code.claude.com/docs/en/hooks diff --git a/uk/INDEX.md b/uk/INDEX.md index 97b5405..ca8bf84 100644 --- a/uk/INDEX.md +++ b/uk/INDEX.md @@ -237,7 +237,7 @@ blog-draft/ - Хуки інструментів: PreToolUse, PostToolUse, PostToolUseFailure, PostToolBatch, PermissionRequest, PermissionDenied - Хуки сесії: SessionStart, Setup, SessionEnd, Stop, StopFailure, SubagentStart, SubagentStop -- Хуки завдань: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle +- Хуки завдань: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle (TaskCompleted/TaskCreated спрацьовують лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) - Хуки життєвого циклу: ConfigChange, CwdChanged, DirectoryAdded, FileChanged, PreCompact, PostCompact, WorktreeCreate, WorktreeRemove, Notification, InstructionsLoaded, Elicitation, ElicitationResult --- @@ -932,10 +932,12 @@ Run tests in background --- -**Останнє оновлення**: 9 квітня 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 **Прикладів**: 100+ файлів **Категорій**: 10 функцій **Хуків**: 8 скриптів автоматизації **Прикладів конфігурації**: 10+ сценаріїв **Готові до використання**: Усі приклади +**Джерела**: +- https://code.claude.com/docs/en/hooks diff --git a/uk/README.md b/uk/README.md index 8a4afb0..b73dcde 100644 --- a/uk/README.md +++ b/uk/README.md @@ -443,7 +443,7 @@ chmod +x ~/.claude/hooks/*.sh **Події хуків** (31, у 4 категоріях) — коли він виконується: - **Хуки інструментів**: `PreToolUse`, `PostToolUse`, `PostToolUseFailure`, `PostToolBatch`, `PermissionRequest`, `PermissionDenied` - **Хуки сесій**: `SessionStart`, `Setup`, `SessionEnd`, `Stop`, `StopFailure`, `SubagentStart`, `SubagentStop` -- **Хуки завдань**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle` +- **Хуки завдань**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle` (`TaskCompleted` і `TaskCreated` спрацьовують лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) - **Хуки життєвого циклу**: `ConfigChange`, `CwdChanged`, `DirectoryAdded`, `FileChanged`, `PreCompact`, `PostCompact`, `WorktreeCreate`, `WorktreeRemove`, `Notification`, `InstructionsLoaded`, `Elicitation`, `ElicitationResult` @@ -659,6 +659,8 @@ Claude: --- -**Останнє оновлення**: Квітень 2026 -**Версія Claude Code**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/hooks **Сумісні моделі**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/uk/STYLE_GUIDE.md b/uk/STYLE_GUIDE.md index 63a4b8b..27f2437 100644 --- a/uk/STYLE_GUIDE.md +++ b/uk/STYLE_GUIDE.md @@ -634,4 +634,8 @@ README уроків завершуються блоком метаданих: - [ ] Повідомлення коміту відповідає формату conventional commits --- -**Останнє оновлення**: Квітень 2026 + +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/overview diff --git a/uk/claude_concepts_guide.md b/uk/claude_concepts_guide.md index b3a5b30..fb8c998 100644 --- a/uk/claude_concepts_guide.md +++ b/uk/claude_concepts_guide.md @@ -2826,8 +2826,8 @@ Claude Code підтримує **31 подія хуків** у п'яти тип | **Stop** | Claude завершив відповідь | Генерація підсумку, завдання очищення | | **StopFailure** | Помилка API завершує хід | Відновлення після помилки, логування | | **TeammateIdle** | Тімейт у команді агентів без роботи | Розподіл роботи, координація | -| **TaskCompleted** | Завдання позначено як виконане | Пост-обробка завдання | -| **TaskCreated** | Завдання створено через TaskCreate | Відстеження завдань, логування | +| **TaskCompleted** | Завдання позначено як виконане (спрацьовує лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) | Пост-обробка завдання | +| **TaskCreated** | Завдання створено через TaskCreate (спрацьовує лише коли увімкнені todo-інструменти — типово вимкнені на Opus 4.8, Sonnet 5, Fable 5, Mythos 5 і новіших; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` повертає їх) | Відстеження завдань, логування | | **ConfigChange** | Зміна конфігураційного файлу | Валідація, поширення | | **CwdChanged** | Зміна робочого каталогу | Налаштування для конкретного каталогу | | **FileChanged** | Зміна відстежуваного файлу | Моніторинг файлів, тригери перебудови | @@ -3127,10 +3127,12 @@ User: \ --- -*Останнє оновлення: квітень 2026* +*Останнє оновлення: 15 серпня 2026* *Для Claude Haiku 4.5, Sonnet 4.6 та Opus 4.6* *Тепер включає: хуки, контрольні точки, режим планування, розширене мислення, фонові завдання, режими дозволів (6 режимів), headless-режим, управління сесіями, авто-пам'ять, команди агентів, заплановані завдання, інтеграцію з Chrome, канали, голосовий ввід та комплектні навички* --- -**Last Updated**: April 9, 2026 -**Claude Code Version**: 2.1.97 +**Останнє оновлення**: 15 серпня 2026 +**Версія Claude Code**: 2.1.233 +**Джерела**: +- https://code.claude.com/docs/en/hooks diff --git a/vi/01-slash-commands/README.md b/vi/01-slash-commands/README.md index 6843a97..e82c8e6 100644 --- a/vi/01-slash-commands/README.md +++ b/vi/01-slash-commands/README.md @@ -66,12 +66,12 @@ Lệnh tích hợp sẵn là các lối tắt cho các hành động phổ biế | `/powerup` | Khám phá tính năng thông qua các bài học tương tác với demo hoạt hình | | `/privacy-settings` | Cài đặt quyền riêng tư (chỉ Pro/Max) | | `/release-notes` | Xem changelog | -| `/reload-plugins` | Tải lại các plugins hoạt động | +| `/reload-plugins` | Tải lại các plugins hoạt động. Từ v2.1.221, hầu hết các lần cài đặt đều kích hoạt ngay, nên chỉ cần lệnh này khi bản tóm tắt cài đặt hiển thị `Run /reload-plugins to activate.` | | `/remote-control` | Điều khiển từ xa từ claude.ai (bí danh: `/rc`) | | `/remote-env` | Cấu hình môi trường từ xa mặc định | | `/rename [name]` | Đổi tên phiên | | `/resume [session]` | Tiếp tục cuộc hội thoại (bí danh: `/continue`) | -| `/review` | **Đã lỗi thời** — cài đặt plugin `code-review` thay thế | +| `/review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | Bí danh của `/code-review` (v2.1.223): review diff hiện tại, hoặc số PR, nhánh, hay đường dẫn bạn truyền vào — ví dụ `/review 1234`. Nhận cùng các mức effort và flags. Nếu không chỉ định mức, nó dùng lại mức `low`–`max` bạn đã gõ lần trước | | `/rewind` | Quay lại cuộc hội thoại và/hoặc code (bí danh: `/checkpoint`) | | `/sandbox` | Bật/tắt chế độ sandbox | | `/schedule [description]` | Tạo/quản lý các tác vụ định kỳ | @@ -84,7 +84,6 @@ Lệnh tích hợp sẵn là các lối tắt cho các hành động phổ biế | `/tasks` | Liệt kê/quản lý các tác vụ nền | | `/terminal-setup` | Cấu hình terminal keybindings | | `/theme` | Thay đổi chủ đề màu | -| `/ultraplan ` | Soạn kế hoạch trong ultraplan session, xem trong trình duyệt | | `/upgrade` | Mở trang nâng cấp cho tier cao hơn | | `/voice` | Bật/tắt nhập liệu giọng nói push-to-talk | @@ -104,7 +103,6 @@ Những skills này được gửi kèm với Claude Code và được gọi nh | Lệnh | Trạng Thái | |---------|--------| -| `/review` | Đã lỗi thời — được thay thế bởi plugin `code-review` | | `/output-style` | Đã xóa trong v2.1.91 (lỗi thời từ v2.1.73) — dùng `/config` → Output style, hoặc setting `outputStyle` | | `/pr-comments` | Đã xóa trong v2.1.91 — hỏi Claude trực tiếp để xem bình luận PR | | `/vim` | Đã xóa trong v2.1.92 — sử dụng /config → Editor mode | @@ -113,14 +111,14 @@ Những skills này được gửi kèm với Claude Code và được gọi nh - `/fork` và `/subtask` hoán đổi vai trò trong **v2.1.212**. `/fork` giờ sao chép cuộc hội thoại thành một phiên nền độc lập mới; hành vi subagent-được-fork mà nó từng có đã chuyển sang lệnh mới `/subtask`. Lịch sử: `/fork` là bí danh của `/branch` từ v2.1.77 đến v2.1.161; từ v2.1.161 đến v2.1.211 nó khởi chạy một subagent được fork (việc mà `/subtask` làm bây giờ). Khi tắt agent view, `/subtask` không khả dụng và `/fork` giữ hành vi subagent-được-fork - `/output-style` đã lỗi thời (v2.1.73) và bị xóa (v2.1.91) — output styles vẫn có sẵn qua `/config` → Output style hoặc setting `outputStyle`; các bản tích hợp sẵn là Default, Proactive, Explanatory, và Learning -- `/review` đã lỗi thời thay vào đó là plugin `code-review` +- `/review` trở thành bí danh đầy đủ của `/code-review` — cùng target, mức effort và flags (v2.1.223). Lịch sử: ban đầu nó chuyển sang engine `/code-review medium` trong v2.1.186 nhưng vẫn chỉ dùng cho PR - Lệnh `/effort` được thêm với mức `max` yêu cầu Opus 4.6 - Lệnh `/voice` được thêm cho nhập liệu giọng nói push-to-talk - Lệnh `/schedule` được thêm để tạo/quản lý các tác vụ định kỳ - Lệnh `/color` được thêm để tùy chỉnh thanh prompt - `/pr-comments` đã xóa trong v2.1.91 — hỏi Claude trực tiếp để xem bình luận PR - `/vim` đã xóa trong v2.1.92 — sử dụng /config → Editor mode thay thế -- `/ultraplan` được thêm để xem và thực thi kế hoạch trong trình duyệt +- `/ultraplan` đã bị gỡ bỏ trong v2.1.222 — hãy dùng plan mode thay thế - `/powerup` được thêm để học tính năng tương tác - `/sandbox` được thêm để bật/tắt chế độ sandbox - Bộ chọn `/model` hiện hiển thị nhãn dễ đọc cho con người (ví dụ: "Sonnet 4.6") thay vì ID mô hình thô @@ -559,8 +557,10 @@ Nếu cả hai tồn tại với cùng tên, **skill sẽ được ưu tiên**. --- -**Cập Nhật Lần Cuối**: Tháng 4 năm 2026 -**Phiên Bản Claude Code**: 2.1+ +**Cập Nhật Lần Cuối**: Ngày 15 tháng 8 năm 2026 +**Phiên Bản Claude Code**: 2.1.233 +**Nguồn**: +- https://code.claude.com/docs/en/commands **Các Mô Hình Tương Thích**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 *Phần của series hướng dẫn [Claude How To](../)* diff --git a/vi/06-hooks/README.md b/vi/06-hooks/README.md index 822e285..1b01916 100644 --- a/vi/06-hooks/README.md +++ b/vi/06-hooks/README.md @@ -201,6 +201,8 @@ Claude Code hỗ trợ **31 sự kiện hook**: | **ElicitationResult** | Người dùng phản hồi elicitation | (none) | Có | Xử lý phản hồi | | **SessionEnd** | Phiên kết thúc | (none) | Không | Dọn dẹp, logging cuối | +> **`TaskCreated` và `TaskCompleted` cần bật todo tools (v2.1.233).** Hai sự kiện này phát ra từ các todo/task-tracking tools (`TaskCreate`/`Get`/`Update`/`List`, `TodoWrite`), vốn **không còn khả dụng trên Opus 4.8, Sonnet 5, Fable 5, Mythos 5 và các model mới hơn**. Trên các model đó, hook vẫn là cấu hình hợp lệ nhưng đơn giản là không bao giờ chạy — không có output và cũng không có lỗi. Đặt `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` để khôi phục. + ### PreToolUse Chạy sau khi Claude tạo tham số công cụ và trước khi xử lý. Sử dụng điều này để xác thực hoặc sửa đổi đầu vào công cụ. @@ -864,6 +866,8 @@ Chỉnh sửa `~/.claude/settings.json` hoặc `.claude/settings.json` với c --- -**Cập Nhật Lần Cuối**: Tháng 4 năm 2026 -**Phiên Bản Claude Code**: 2.1+ +**Cập Nhật Lần Cuối**: Ngày 15 tháng 8 năm 2026 +**Phiên Bản Claude Code**: 2.1.233 +**Nguồn**: +- https://code.claude.com/docs/en/hooks **Các Mô Hình Tương Thích**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/vi/07-plugins/README.md b/vi/07-plugins/README.md index f78c61b..7ec7d67 100644 --- a/vi/07-plugins/README.md +++ b/vi/07-plugins/README.md @@ -457,6 +457,10 @@ Người dùng enterprise và nâng cao có thể kiểm soát hành vi marketpl | `strictKnownMarketplaces` | Kiểm soát marketplace nào người dùng được phép thêm | | `deniedPlugins` | Blocklist do quản trị viên quản lý để ngăn cài đặt các plugin cụ thể | +> **Bí danh thân thiện hơn (v2.1.232)**: `additionalMarketplaces` được chấp nhận như bí danh của `extraKnownMarketplaces`, và `allowedMarketplaces` của `strictKnownMarketplaces`. **Nguồn changelog** — changelog v2.1.232 công bố chúng, nhưng trang settings chính thức chưa liệt kê. Các khóa chuẩn vẫn an toàn để tiếp tục dùng. + +> **Ký tự đại diện theo owner (v2.1.223+)**: một mục `"owner/*"` cho phép hoặc chặn mọi repo marketplace thuộc một owner GitHub. **Chỉ được chấp nhận trong `strictKnownMarketplaces` và `blockedMarketplaces`.** Ở mọi nơi khác dùng nguồn `github` — kể cả `extraKnownMarketplaces` và `/plugin marketplace add` — giá trị `repo` phải chỉ đúng một repository. + ### Các Tính Năng Marketplace Thêm / Additional Marketplace Features - **Git timeout mặc định**: Tăng từ 30s lên 120s cho các repository plugin lớn @@ -616,6 +620,15 @@ claude plugin validate # Xác thực cấu trúc plugin claude plugin install plugin-name@marketplace-name ``` +**Có hiệu lực ngay không?** Từ **v2.1.221**, thường là có — hãy đọc dòng cuối của bản tóm tắt cài đặt: + +| Bản tóm tắt cài đặt hiển thị | Ý nghĩa | +|---|---| +| `Plugin is now active.` | Claude Code đã kích hoạt plugin ngay trong quá trình cài đặt. Không cần làm gì thêm. | +| `Run /reload-plugins to activate.` | Plugin đã được cài nhưng chưa hoạt động — hoặc vì việc kích hoạt sẽ làm mất hiệu lực prompt cache, hoặc vì lần kích hoạt đã thất bại. | + +Trước v2.1.221, không lần cài đặt nào có hiệu lực trong phiên hiện tại cho đến khi bạn chạy `/reload-plugins` hoặc khởi động lại. + ### Kích Hoạt / Vô Hiệu Hóa (với phạm vi tự phát hiện) ```bash /plugin enable plugin-name @@ -701,6 +714,10 @@ Các quản trị viên có thể kiểm soát hành vi plugin trên một tổ | `strictKnownMarketplaces` | Hạn chế marketplace nào người dùng được phép thêm | | `allowedChannelPlugins` | Kiểm soát plugins nào được phép theo kênh phát hành | +> **Bí danh thân thiện hơn (v2.1.232)**: `additionalMarketplaces` được chấp nhận như bí danh của `extraKnownMarketplaces`, và `allowedMarketplaces` của `strictKnownMarketplaces`. **Nguồn changelog** — changelog v2.1.232 công bố chúng, nhưng trang settings chính thức chưa liệt kê. Các khóa chuẩn vẫn an toàn để tiếp tục dùng. + +> **Ký tự đại diện theo owner (v2.1.223+)**: một mục `"owner/*"` cho phép hoặc chặn mọi repo marketplace thuộc một owner GitHub. **Chỉ được chấp nhận trong `strictKnownMarketplaces` và `blockedMarketplaces`.** Ở mọi nơi khác dùng nguồn `github` — kể cả `extraKnownMarketplaces` và `/plugin marketplace add` — giá trị `repo` phải chỉ đúng một repository. + Các cài đặt này có thể được áp dụng ở cấp tổ chức qua các file cấu hình được quản lý và có ưu tiên hơn các cài đặt cấp người dùng. ## Bảo Mật Plugin / Plugin Security @@ -944,6 +961,9 @@ Các tính năng Claude Code sau hoạt động cùng với plugins: --- -**Cập Nhật Lần Cuối**: Tháng 4 năm 2026 -**Phiên Bản Claude Code**: 2.1+ +**Cập Nhật Lần Cuối**: Ngày 15 tháng 8 năm 2026 +**Phiên Bản Claude Code**: 2.1.233 +**Nguồn**: +- https://code.claude.com/docs/en/discover-plugins +- https://code.claude.com/docs/en/settings **Các Mô Hình Tương Thích**: Claude Sonnet 4.6, Claude Opus 4.6, Claude Haiku 4.5 diff --git a/vi/CATALOG.md b/vi/CATALOG.md index 4ee6991..4ad0445 100644 --- a/vi/CATALOG.md +++ b/vi/CATALOG.md @@ -65,7 +65,7 @@ Commands là các lệnh tắt do người dùng gọi thực hiện các hành | `/sandbox` | Bật/tắt sandbox mode | Thực thi command an toàn | | `/vim` | Bật/tắt vim mode | Chỉnh sửa kiểu Vim | | `/doctor` | Chạy chẩn đoán | Khắc phục sự cố | -| `/reload-plugins` | Tải lại plugins đã cài đặt | Quản lý plugin | +| `/reload-plugins` | Tải lại plugins đã cài đặt. Từ v2.1.221, hầu hết các lần cài đặt kích hoạt ngay; chỉ cần khi bản tóm tắt cài đặt hiển thị `Run /reload-plugins to activate.` | Quản lý plugin | | `/release-notes` | Hiển thị release notes | Kiểm tra tính năng mới | | `/remote-control` | Bật điều khiển từ xa | Truy cập từ xa | | `/permissions` | Quản lý permissions | Kiểm soát truy cập | @@ -344,8 +344,8 @@ Tự động hóa dựa trên sự kiện thực thi shell commands trên các s | `Stop` | Claude hoàn thành phản hồi | Phản hồi hoàn tất | Dọn dẹp, báo cáo | | `StopFailure` | Lỗi API kết thúc turn | Lỗi API xảy ra | Phục hồi lỗi, logging | | `TeammateIdle` | Agent teammate rảnh | Phối hợp agent team | Phân phối work | -| `TaskCompleted` | Task được đánh dấu hoàn tất | Task xong | Xử lý post-task | -| `TaskCreated` | Task được tạo qua TaskCreate | Task mới được tạo | Theo dõi task, logging | +| `TaskCompleted` | Task được đánh dấu hoàn tất (chỉ phát ra khi bật todo tools — tắt mặc định trên Opus 4.8, Sonnet 5, Fable 5, Mythos 5 và mới hơn; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` khôi phục) | Task xong | Xử lý post-task | +| `TaskCreated` | Task được tạo qua TaskCreate (chỉ phát ra khi bật todo tools — tắt mặc định trên Opus 4.8, Sonnet 5, Fable 5, Mythos 5 và mới hơn; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` khôi phục) | Task mới được tạo | Theo dõi task, logging | | `ConfigChange` | Cấu hình được cập nhật | Settings được sửa đổi | Phản ứng thay đổi config | | `CwdChanged` | Thư mục làm việc thay đổi | Thư mục thay đổi | Thiết lập cụ thể theo thư mục | | `FileChanged` | File được theo dõi thay đổi | File được sửa | Giám sát file, rebuild | @@ -514,4 +514,8 @@ chmod +x ~/.claude/hooks/*.sh --- -**Cập Nhật Lần**: Tháng 3 năm 2026 +**Cập Nhật Lần Cuối**: Ngày 15 tháng 8 năm 2026 +**Phiên Bản Claude Code**: 2.1.233 +**Nguồn**: +- https://code.claude.com/docs/en/commands +- https://code.claude.com/docs/en/hooks diff --git a/vi/INDEX.md b/vi/INDEX.md index ca10b38..7c14218 100644 --- a/vi/INDEX.md +++ b/vi/INDEX.md @@ -223,7 +223,7 @@ Các script tự động hóa dựa trên sự kiện thực thi tự động. **Hook Events** (31, trong 4 nhóm) — khi nào hook chạy: - Tool Hooks: PreToolUse, PostToolUse, PostToolUseFailure, PostToolBatch, PermissionRequest, PermissionDenied - Session Hooks: SessionStart, Setup, SessionEnd, Stop, StopFailure, SubagentStart, SubagentStop -- Task Hooks: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle +- Task Hooks: UserPromptSubmit, UserPromptExpansion, MessageDisplay, TaskCompleted, TaskCreated, TeammateIdle (TaskCompleted/TaskCreated chỉ phát ra khi bật todo tools — tắt mặc định trên Opus 4.8, Sonnet 5, Fable 5, Mythos 5 và mới hơn; `CLAUDE_CODE_ENABLE_TODO_TOOLS=1` khôi phục) - Lifecycle Hooks: ConfigChange, CwdChanged, DirectoryAdded, FileChanged, PreCompact, PostCompact, WorktreeCreate, WorktreeRemove, Notification, InstructionsLoaded, Elicitation, ElicitationResult --- @@ -876,9 +876,12 @@ Muốn thêm ví dụ nữa? Theo cấu trúc: --- -**Cập Nhật Lần**: Tháng 3 năm 2026 +**Cập Nhật Lần Cuối**: Ngày 15 tháng 8 năm 2026 +**Phiên Bản Claude Code**: 2.1.233 **Tổng Số Ví Dụ**: 100+ files **Danh Mục**: 10 tính năng **Hooks**: 8 scripts tự động hóa **Ví Dụ Cấu Hình**: 10+ scenarios **Sẵn Sàng Sử Dụng**: Tất cả ví dụ +**Nguồn**: +- https://code.claude.com/docs/en/hooks diff --git a/zh/01-slash-commands/README.md b/zh/01-slash-commands/README.md index c92b04e..10084db 100644 --- a/zh/01-slash-commands/README.md +++ b/zh/01-slash-commands/README.md @@ -66,12 +66,12 @@ Claude Code 目前提供 55+ 个内置命令和 5 个内置 Skills。你可以 | `/pr-comments [PR]` | 获取 GitHub PR 评论 | | `/privacy-settings` | 隐私设置（仅 Pro/Max） | | `/release-notes` | 查看更新日志 | -| `/reload-plugins` | 重新加载当前插件 | +| `/reload-plugins` | 重新加载当前插件。自 v2.1.221 起，大多数安装会立即生效，只有当安装摘要提示 `Run /reload-plugins to activate.` 时才需要执行 | | `/remote-control` | 从 claude.ai 进行远程控制（别名：`/rc`） | | `/remote-env` | 配置默认远程环境 | | `/rename [name]` | 重命名会话 | | `/resume [session]` | 恢复对话（别名：`/continue`） | -| `/review` | **已弃用**，请改用 `code-review` 插件 | +| `/review [low\|medium\|high\|xhigh\|max\|ultra] [--fix] [--comment] [pr#\|branch\|path]` | `/code-review` 的别名（v2.1.223）：审查当前 diff，或你传入的 PR 编号、分支、路径 —— 例如 `/review 1234`。接受相同的 effort 级别和标志。未指定级别时，复用你上次输入的 `low`–`max` 级别 | | `/rewind` | 回退对话和/或代码（别名：`/checkpoint`） | | `/sandbox` | 切换沙盒模式 | | `/schedule [description]` | 创建/管理定时任务 | @@ -101,7 +101,6 @@ Claude Code 目前提供 55+ 个内置命令和 5 个内置 Skills。你可以 | 命令 | 状态 | |---------|--------| -| `/review` | 已弃用，已被 `code-review` 插件替代 | | `/output-style` | 已于 v2.1.91 移除（v2.1.73 起弃用）— 改用 `/config` → Output style，或 `outputStyle` 设置 | | `/vim` | 自 v2.1.92 起移除；改用 `/config → Editor mode` | @@ -109,7 +108,7 @@ Claude Code 目前提供 55+ 个内置命令和 5 个内置 Skills。你可以 - `/fork` 与 `/subtask` 在 **v2.1.212** 互换了角色。`/fork` 现在是把对话复制到一个独立的后台会话；原来的 forked subagent 行为转移到了新命令 `/subtask`。历史：v2.1.77–v2.1.161 期间 `/fork` 是 `/branch` 的别名；v2.1.161–v2.1.211 期间它启动 forked subagent（即现在 `/subtask` 的行为）。关闭 agent view 时 `/subtask` 不可用，`/fork` 仍保留 forked subagent 行为 - `/output-style` 已弃用（v2.1.73）并于 v2.1.91 移除 — output styles 仍可通过 `/config` → Output style 或 `outputStyle` 设置使用 -- `/review` 已弃用，推荐改用 `code-review` 插件 +- `/review` 成为 `/code-review` 的完整别名 —— 目标、effort 级别和标志完全相同（v2.1.223）。历史：它在 v2.1.186 首次迁移到 `/code-review medium` 引擎，但当时仅支持 PR - 新增 `/effort`，其中 `max` 级别需要 Opus 4.6 - 新增 `/voice`，用于按住说话语音输入 - 新增 `/schedule`，用于创建和管理定时任务 @@ -549,8 +548,10 @@ allowed-tools: Bash(npm *), Bash(git *) --- -**最后更新**: 2026 年 4 月 9 日 -**Claude Code 版本**: 2.1.97 +**最后更新**: 2026 年 8 月 15 日 +**Claude Code 版本**: 2.1.233 +**来源**: +- https://code.claude.com/docs/en/commands --- diff --git a/zh/06-hooks/README.md b/zh/06-hooks/README.md index 2b6a50f..2a23092 100644 --- a/zh/06-hooks/README.md +++ b/zh/06-hooks/README.md @@ -26,7 +26,7 @@ Claude Code 支持 31 个 hook 事件，按用途分为以下 4 组。事件的 - **Tool Hooks**：`PreToolUse`、`PostToolUse`、`PostToolUseFailure`、`PostToolBatch`、`PermissionRequest`、`PermissionDenied` - **Session Hooks**：`SessionStart`、`Setup`、`SessionEnd`、`Stop`、`StopFailure`、`SubagentStart`、`SubagentStop` -- **Task Hooks**：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle` +- **Task Hooks**：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle`（`TaskCompleted` 和 `TaskCreated` 仅在启用 todo 工具时触发 —— 在 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 及更新模型上默认关闭；`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` 可恢复） - **Lifecycle Hooks**：`ConfigChange`、`CwdChanged`、`DirectoryAdded`、`FileChanged`、`PreCompact`、`PostCompact`、`WorktreeCreate`、`WorktreeRemove`、`Notification`、`InstructionsLoaded`、`Elicitation`、`ElicitationResult` ## 安装 @@ -110,3 +110,10 @@ Hooks 会在匹配到事件时自动执行。你可以把它理解成 Claude Cod - [Memory Guide](../02-memory/README.md) - [Official Hooks Documentation](https://code.claude.com/docs/en/hooks) - [CLI Reference](https://code.claude.com/docs/en/cli-reference) + +--- + +**最后更新**: 2026 年 8 月 15 日 +**Claude Code 版本**: 2.1.233 +**来源**: +- https://code.claude.com/docs/en/hooks diff --git a/zh/07-plugins/README.md b/zh/07-plugins/README.md index 4df21c2..b260a82 100644 --- a/zh/07-plugins/README.md +++ b/zh/07-plugins/README.md @@ -444,6 +444,10 @@ graph TB | `strictKnownMarketplaces` | 控制允许用户添加哪些市场 | | `deniedPlugins` | 管理员维护的黑名单，阻止特定插件被安装 | +> **更友好的别名（v2.1.232）**：`additionalMarketplaces` 被接受为 `extraKnownMarketplaces` 的别名，`allowedMarketplaces` 为 `strictKnownMarketplaces` 的别名。**来源为 changelog** —— v2.1.232 的 changelog 公布了它们，但官方 settings 参考页面尚未列出。继续使用规范键名是安全的。 + +> **owner 通配符（v2.1.223+）**：`"owner/*"` 条目可允许或阻止某个 GitHub owner 名下的所有市场仓库。**仅在 `strictKnownMarketplaces` 和 `blockedMarketplaces` 中被接受。** 在其他出现 `github` 源的地方 —— 包括 `extraKnownMarketplaces` 和 `/plugin marketplace add` —— `repo` 的值必须指向单个仓库。 + ### 额外的市场特性 - **默认 git 超时**：对大型插件仓库从 30 秒增加到 120 秒 @@ -603,6 +607,15 @@ claude plugin validate # 验证插件结构 claude plugin install plugin-name@marketplace-name ``` +**是否立即生效？** 自 **v2.1.221** 起通常会立即生效 —— 请查看安装摘要的最后一行： + +| 安装摘要显示 | 含义 | +|---|---| +| `Plugin is now active.` | Claude Code 已在安装过程中激活该插件，无需其他操作。 | +| `Run /reload-plugins to activate.` | 插件已安装但尚未生效 —— 要么激活会使 prompt cache 失效，要么激活尝试失败。 | + +在 v2.1.221 之前，任何安装都要等到你运行 `/reload-plugins` 或重启后才会在当前会话中生效。 + ### 启用 / 禁用（自动检测作用域） ```bash /plugin enable plugin-name @@ -688,6 +701,10 @@ claude --plugin-dir ./my-plugin --plugin-dir ./another-plugin | `strictKnownMarketplaces` | 限制用户允许添加的市场 | | `allowedChannelPlugins` | 控制每个发布渠道允许使用哪些插件 | +> **更友好的别名（v2.1.232）**：`additionalMarketplaces` 被接受为 `extraKnownMarketplaces` 的别名，`allowedMarketplaces` 为 `strictKnownMarketplaces` 的别名。**来源为 changelog** —— v2.1.232 的 changelog 公布了它们，但官方 settings 参考页面尚未列出。继续使用规范键名是安全的。 + +> **owner 通配符（v2.1.223+）**：`"owner/*"` 条目可允许或阻止某个 GitHub owner 名下的所有市场仓库。**仅在 `strictKnownMarketplaces` 和 `blockedMarketplaces` 中被接受。** 在其他出现 `github` 源的地方 —— 包括 `extraKnownMarketplaces` 和 `/plugin marketplace add` —— `repo` 的值必须指向单个仓库。 + 这些设置可以通过托管配置文件应用到组织级别，并且优先于用户级设置。 ## 插件安全 @@ -928,3 +945,11 @@ claude --plugin-dir ./my-plugin --plugin-dir ./another-plugin - [MCP Server 参考](https://modelcontextprotocol.io/) - [Subagent 配置指南](../04-subagents/README.md) - [Hook 系统参考](../06-hooks/README.md) + +--- + +**最后更新**: 2026 年 8 月 15 日 +**Claude Code 版本**: 2.1.233 +**来源**: +- https://code.claude.com/docs/en/discover-plugins +- https://code.claude.com/docs/en/settings diff --git a/zh/CATALOG.md b/zh/CATALOG.md index 1c4b2e4..24203ad 100644 --- a/zh/CATALOG.md +++ b/zh/CATALOG.md @@ -64,7 +64,7 @@ | `/logout` | 登出 | 切换账号 | | `/sandbox` | 切换 sandbox 模式 | 安全执行命令 | | `/doctor` | 运行诊断 | 排查问题 | -| `/reload-plugins` | 重新加载已安装的 plugins | 插件管理 | +| `/reload-plugins` | 重新加载已安装的 plugins。自 v2.1.221 起，大多数安装会立即生效；仅当安装摘要提示 `Run /reload-plugins to activate.` 时才需要 | 插件管理 | | `/release-notes` | 显示更新说明 | 查看新功能 | | `/remote-control` | 启用远程控制 | 远程访问 | | `/permissions` | 管理权限 | 控制访问 | @@ -348,8 +348,8 @@ export GITHUB_TOKEN="your_token" && claude mcp add github -- npx -y @modelcontex | `Stop` | Claude 完成响应 | 响应完成 | 清理、汇报 | | `StopFailure` | API 错误导致结束 | API 错误发生 | 错误恢复、日志 | | `TeammateIdle` | 队友 agent 空闲 | agent team 协调 | 分配工作 | -| `TaskCompleted` | 任务标记完成 | 任务完成 | 任务后处理 | -| `TaskCreated` | 通过 TaskCreate 创建任务 | 新任务创建 | 任务追踪、日志 | +| `TaskCompleted` | 任务标记完成（仅在启用 todo 工具时触发 —— 在 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 及更新模型上默认关闭；`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` 可恢复） | 任务完成 | 任务后处理 | +| `TaskCreated` | 通过 TaskCreate 创建任务（仅在启用 todo 工具时触发 —— 在 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 及更新模型上默认关闭；`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` 可恢复） | 新任务创建 | 任务追踪、日志 | | `ConfigChange` | 配置更新 | 设置被修改 | 响应配置变化 | | `CwdChanged` | 当前工作目录变化 | 目录切换 | 目录级初始化 | | `FileChanged` | 监控文件发生变化 | 文件被修改 | 文件监控、重建 | @@ -523,5 +523,8 @@ claude mcp add github -- npx -y @modelcontextprotocol/server-github --- -**最后更新**: 2026 年 4 月 9 日 -**Claude Code 版本**: 2.1.97 +**最后更新**: 2026 年 8 月 15 日 +**Claude Code 版本**: 2.1.233 +**来源**: +- https://code.claude.com/docs/en/commands +- https://code.claude.com/docs/en/hooks diff --git a/zh/INDEX.md b/zh/INDEX.md index 4a94430..33c4d33 100644 --- a/zh/INDEX.md +++ b/zh/INDEX.md @@ -219,7 +219,7 @@ blog-draft/ **Hook 事件**（31 个，分 4 类）— 决定 hook 何时运行： - 工具 Hook：`PreToolUse`、`PostToolUse`、`PostToolUseFailure`、`PostToolBatch`、`PermissionRequest`、`PermissionDenied` - 会话 Hook：`SessionStart`、`Setup`、`SessionEnd`、`Stop`、`StopFailure`、`SubagentStart`、`SubagentStop` -- 任务 Hook：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle` +- 任务 Hook：`UserPromptSubmit`、`UserPromptExpansion`、`MessageDisplay`、`TaskCompleted`、`TaskCreated`、`TeammateIdle`（TaskCompleted/TaskCreated 仅在启用 todo 工具时触发 —— 在 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 及更新模型上默认关闭；`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` 可恢复） - 生命周期 Hook：`ConfigChange`、`CwdChanged`、`DirectoryAdded`、`FileChanged`、`PreCompact`、`PostCompact`、`WorktreeCreate`、`WorktreeRemove`、`Notification`、`InstructionsLoaded`、`Elicitation`、`ElicitationResult` --- @@ -894,9 +894,12 @@ claude -p "Run tests and report results" --- -**最后更新**：2026 年 3 月 +**最后更新**：2026 年 8 月 15 日 +**Claude Code 版本**：2.1.233 **总示例数**：100+ 文件 **分类数**：10 个功能 **Hooks**：8 个自动化脚本 **配置示例**：10+ 个场景 **可直接使用**：所有示例 +**来源**： +- https://code.claude.com/docs/en/hooks diff --git a/zh/README.md b/zh/README.md index b65ec8f..b6ce7ca 100644 --- a/zh/README.md +++ b/zh/README.md @@ -483,7 +483,7 @@ chmod +x ~/.claude/hooks/*.sh **Hook 事件**（31 个，分 4 类）— 决定 hook 何时运行: - **工具 Hook**: `PreToolUse`, `PostToolUse`, `PostToolUseFailure`, `PostToolBatch`, `PermissionRequest`, `PermissionDenied` - **会话 Hook**: `SessionStart`, `Setup`, `SessionEnd`, `Stop`, `StopFailure`, `SubagentStart`, `SubagentStop` -- **任务 Hook**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle` +- **任务 Hook**: `UserPromptSubmit`, `UserPromptExpansion`, `MessageDisplay`, `TaskCompleted`, `TaskCreated`, `TeammateIdle`（`TaskCompleted` 和 `TaskCreated` 仅在启用 todo 工具时触发 —— 在 Opus 4.8、Sonnet 5、Fable 5、Mythos 5 及更新模型上默认关闭；`CLAUDE_CODE_ENABLE_TODO_TOOLS=1` 可恢复） - **生命周期 Hook**: `ConfigChange`, `CwdChanged`, `DirectoryAdded`, `FileChanged`, `PreCompact`, `PostCompact`, `WorktreeCreate`, `WorktreeRemove`, `Notification`, `InstructionsLoaded`, `Elicitation`, `ElicitationResult` @@ -878,6 +878,8 @@ MIT 许可证，详见 [LICENSE](LICENSE)。你可以自由使用、修改和分 --- -**最后更新**：2026 年 3 月 -**Claude Code 版本**：2.1+ +**最后更新**：2026 年 8 月 15 日 +**Claude Code 版本**：2.1.233 +**来源**： +- https://code.claude.com/docs/en/hooks **兼容模型**：Claude Sonnet 4.6、Claude Opus 4.6、Claude Haiku 4.5 diff --git a/zh/STYLE_GUIDE.md b/zh/STYLE_GUIDE.md index 6aae930..d7e0ad5 100644 --- a/zh/STYLE_GUIDE.md +++ b/zh/STYLE_GUIDE.md @@ -628,3 +628,10 @@ docs(skills): Add comprehensive code review skill - [ ] 段落简短且聚焦 - [ ] 相关指南章节链接到合适的课程 - [ ] 提交信息符合 conventional commits 格式 + +--- + +**最后更新**: 2026 年 8 月 15 日 +**Claude Code 版本**: 2.1.233 +**来源**: +- https://code.claude.com/docs/en/overview