Loading...
Loading...
Compare original and translation side by side
settings.jsonCLAUDE.mdrules/skills/agents/commands/settings.jsonCLAUDE.mdrules/skills/agents/commands/ghgh api repos/anthropics/claude-code/contents/CHANGELOG.md \
--jq '.content' | base64 -d > .i9wa4/tmp/claude-code-changelog.mdghgh api repos/anthropics/claude-code/contents/CHANGELOG.md \
--jq '.content' | base64 -d > .i9wa4/tmp/claude-code-changelog.md## x.x.x## x.x.xDeprecatedRemovedBreakingChangedDeprecatedRemovedBreakingChangedclaude-code-guideTask tool with subagent_type: claude-code-guideclaude-code-guideTask tool with subagent_type: claude-code-guide| Category | Examples |
|---|---|
| Display | |
| Behavior | |
| Tools | |
| MCP | MCP server configurations |
| Hooks | PreToolUse, PostToolUse, Stop hooks |
| Plans | |
| 分类 | 示例 |
|---|---|
| 显示 | |
| 行为 | |
| 工具 | |
| MCP | MCP服务器配置 |
| Hooks | PreToolUse、PostToolUse、Stop hooks |
| 计划 | |
rules/rules/| Type | Load Timing | Purpose |
|---|---|---|
| CLAUDE.md / rules/ | Full load at startup | Global rules always applied |
| commands/ | Explicit user invocation | Predefined prompts, workflows |
| skills/ | Auto-triggered by conversation | Specialized knowledge |
| agents/ | Delegated via Task tool | Independent context |
| 类型 | 加载时机 | 用途 |
|---|---|---|
| CLAUDE.md / rules/ | 启动时完整加载 | 始终应用的全局规则 |
| commands/ | 用户显式调用 | 预定义提示词、工作流 |
| skills/ | 对话时自动触发 | 专业知识 |
| agents/ | 通过Task工具委托 | 独立上下文 |
Bash(cmd *):*git -C *Bash(cmd *):*git -C *languagemcpToolSearchENABLE_TOOL_SEARCHplansDirectory:**CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMSlanguagemcpToolSearchENABLE_TOOL_SEARCHplansDirectory:**CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMSmemory: user|project|localmemory: user|project|localshowTurnDurationreducedMotionModeshowTurnDurationreducedMotionModepages/debugpagesundefinedundefinedbrew install wgetuvx --from git+https://github.com/laiso/site2skill site2skill <URL> <SKILL_NAME>brew install wgetuvx --from git+https://github.com/laiso/site2skill site2skill <URL> <SKILL_NAME>
Options:
- `--output, -o` - Output directory (default: `.claude/skills`)
- `--skill-output` - Where to save .skill file (default: current directory)
- `--skip-fetch` - Skip download (reuse existing files)
- `--clean` - Remove temporary files after completion
To update existing skill docs, re-run without `--skip-fetch`.
选项:
- `--output, -o` - 输出目录(默认:`.claude/skills`)
- `--skill-output` - .skill文件保存路径(默认:当前目录)
- `--skip-fetch` - 跳过下载(重用现有文件)
- `--clean` - 完成后删除临时文件
要更新现有技能文档,重新运行时不添加`--skip-fetch`参数即可。| Mode | Description |
|---|---|
| Prompts for permission on first use of each tool |
| Auto-accepts file edit permissions for the session |
| Plan Mode: analyze only, no modifications |
| Auto-denies unless pre-approved via allow rules |
| Skips all prompts (use only in isolated environments) |
| 模式 | 描述 |
|---|---|
| 首次使用每个工具时提示获取权限 |
| 会话期间自动接受文件编辑权限 |
| 仅分析,不进行修改 |
| 自动拒绝,除非通过允许规则预先批准 |
| 跳过所有提示(仅在隔离环境中使用) |
{
"permissions": {
"allow": [
"Bash(npm run *)",
"Bash(git commit *)",
"Bash(* --version)",
"Bash(* --help *)"
],
"deny": ["Bash(git push *)"]
}
}*Bash(ls *)ls -lalsof{
"permissions": {
"allow": [
"Bash(npm run *)",
"Bash(git commit *)",
"Bash(* --version)",
"Bash(* --help *)"
],
"deny": ["Bash(git push *)"]
}
}*Bash(ls *)ls -lalsof| Pattern | Meaning | Example |
|---|---|---|
| Absolute path from root | |
| Path from home directory | |
| Relative to settings file | |
| Relative to current directory | |
***| 模式 | 含义 | 示例 |
|---|---|---|
| 从根目录开始的绝对路径 | |
| 从用户主目录开始的路径 | |
| 相对于配置文件的路径 | |
| 相对于当前目录的路径 | |
***{
"permissions": {
"allow": ["mcp__puppeteer__*"],
"deny": ["Task(Explore)"]
}
}{
"permissions": {
"allow": ["mcp__puppeteer__*"],
"deny": ["Task(Explore)"]
}
}| Platform | Path |
|---|---|
| macOS | |
| Linux/WSL | |
| Windows | |
| 平台 | 路径 |
|---|---|
| macOS | |
| Linux/WSL | |
| Windows | |
| Setting | Description |
|---|---|
| Set to "disable" to prevent bypass mode |
| Only managed rules apply |
| Only managed/SDK hooks allowed |
| 配置项 | 描述 |
|---|---|
| 设置为"disable"以防止绕过模式 |
| 仅应用托管规则 |
| 仅允许托管/SDK hooks |
https://code.claude.com/docs/en/hookshttps://code.claude.com/docs/en/permissionshttps://code.claude.com/docs/en/skills#inject-dynamic-contexthttps://vercel.com/blog/agents-mdhttps://code.claude.com/docs/en/hookshttps://code.claude.com/docs/en/permissionshttps://code.claude.com/docs/en/skills#inject-dynamic-contexthttps://vercel.com/blog/agents-md