feat: add DevOps-Agent for operational and system administration tasks

README.md

@@ -2,15 +2,15 @@
 
 An AI-powered autonomous coding assistant built with a **Hub-and-Spoke multi-agent architecture** in Go, backed by a Rust-based code analysis engine.
 
-CodeActor Agent orchestrates multiple specialized agents — Conductor, Repo-Analyst, Coding-Engineer, Chat-Assistant, and Meta-Agent — to autonomously analyze, plan, and execute complex software engineering tasks with self-correction capabilities.
+CodeActor Agent orchestrates multiple specialized agents — Conductor, Repo-Analyst, Coding-Engineer, Chat-Assistant, DevOps-Operator, and Meta-Agent — to autonomously analyze, plan, and execute complex software engineering tasks with self-correction capabilities.
 
 ## Features
 
 ### Multi-Agent System
-- **Hub-and-Spoke Architecture** — Central Conductor delegates tasks to specialized sub-agents (Repo analysis, Code editing, General chat)
+- **Hub-and-Spoke Architecture** — Central Conductor delegates tasks to specialized sub-agents (Repo analysis, Code editing, General chat, DevOps operations)
 - **Meta-Agent** — Autonomous agent designer that creates custom sub-agents at runtime for tasks beyond built-in agents' capabilities
 - **Self-Correction** — `thinking` tool enables agents to analyze errors and recover without blind retries
-- **Agent Disable** — Conditionally exclude sub-agents at startup via `--disable-agents=repo,coding,chat,meta`
+- **Agent Disable** — Conditionally exclude sub-agents at startup via `--disable-agents=repo,coding,chat,meta,devops`
 - **ImplPlan Tool** — Stateful implementation plan document for complex multi-step coding tasks
 
 ### Rich Tool System (17 tools)
@@ -42,8 +42,6 @@ CodeActor Agent orchestrates multiple specialized agents — Conductor, Repo-Ana
 - **Call Graph Analysis** — Function-level call graphs with caller/callee relations, cycle detection, complexity scoring
 - **Semantic Code Search** — Vector embeddings via LanceDB + SQLite cache, OpenAI `text-embedding-3-small`
 - **Code Skeleton/Snippet** — Batch file skeleton extraction and per-function code snippet retrieval
-- **File Watching** — Automatic re-indexing on file changes with 20s debounce
-- **Auto-Launch** — Go binary automatically starts the Rust codebase server as a child process with dynamic port allocation, health-check polling, and cleanup on exit
 
 
 ## Screenshots
@@ -91,6 +89,7 @@ enable_streaming = true
 conductor_max_steps = 30
 coding_max_steps = 50
 repo_max_steps = 30
+devops_max_steps = 15
 meta_max_steps = 30
 meta_retry_count = 5
 lang = "Chinese"
@@ -150,10 +149,11 @@ Server defaults to `localhost:9080`. Override via `--host`/`--port` or `CODECACT
 
 | Agent | Tools | Count |
 |-------|-------|-------|
-| Conductor | `delegate_repo`, `delegate_coding`, `delegate_chat`, `delegate_meta`, `finish`, `read_file`, `search_by_regex`, `list_dir`, `print_dir_tree` | 9 |
+| Conductor | `delegate_repo`, `delegate_coding`, `delegate_chat`, `delegate_devops`, `delegate_meta`, `finish`, `read_file`, `search_by_regex`, `list_dir`, `print_dir_tree` | 10 |
 | CodingAgent | All 17 tools (file ops, search, shell, thinking, impl_plan, micro_agent) | 17 |
 | RepoAgent | `read_file`, `search_by_regex`, `list_dir`, `print_dir_tree`, `semantic_search`, `query_code_skeleton`, `query_code_snippet` | 7 |
-| ChatAgent | None (pure LLM conversation) | 0 |
+| ChatAgent | `micro_agent`, `thinking`, `finish` | 3 |
+| DevOpsAgent | `run_bash`, `read_file`, `list_dir`, `print_dir_tree`, `search_by_regex`, `thinking`, `micro_agent`, `finish` | 8 |
 
 [Full architecture documentation →](docs/ARCHITECTURE.md)
 
@@ -185,6 +185,39 @@ Disable Meta-Agent via startup flag:
 ./codeactor tui --disable-agents=meta
 ```
 
+## DevOps-Agent
+
+The **DevOps-Agent** is the operations and infrastructure specialist — it handles all non-coding operational tasks by executing shell commands, inspecting the file system, and analyzing command output. When the Conductor encounters a task involving system administration, log inspection, process management, or ad-hoc shell commands, it delegates to the DevOps-Agent via `delegate_devops`.
+
+### Capabilities
+
+- **Shell Command Execution** (`run_bash`) — Run any bash command with foreground/background support, danger detection, and workspace boundary checks
+- **File System Inspection** — `read_file`, `list_dir`, `print_dir_tree`, `search_by_regex` for browsing logs, configs, and directories
+- **Self-Correction** — `thinking` tool for analyzing command failures and adjusting approach before retrying
+- **Isolated Analysis** — `micro_agent` for deep reasoning on command output or generating structured reports
+
+### Example use cases
+
+- Check disk usage, memory, and system resources
+- Find all log files modified in the last 24 hours
+- Restart services or check process status
+- Inspect configuration files
+- Run system diagnostics and generate reports
+- Execute ad-hoc shell pipelines for data processing
+
+### Configuration
+
+```toml
+[agent]
+devops_max_steps = 15    # Max LLM steps for DevOps-Agent (default: 15)
+```
+
+Disable DevOps-Agent via startup flag:
+
+```bash
+./codeactor tui --disable-agents=devops
+```
+
 ## Codebase Analysis Engine
 
 The `codeactor-codebase` is a standalone **Rust** service that provides deep code analysis capabilities. It runs as a background HTTP server managed automatically by the Go binary.

README_zh.md

@@ -2,15 +2,15 @@
 
 基于 Go 语言开发的 **Hub-and-Spoke（中枢-辐条）多智能体架构** AI 自主编程助手，后端由 Rust 代码分析引擎支撑。
 
-CodeActor Agent 协调多个专用智能体——指挥家（Conductor）、仓库分析员（Repo-Analyst）、编码工程师（Coding-Engineer）、对话助手（Chat-Assistant）和元代理（Meta-Agent）——自主完成复杂的软件工程任务，具备自我纠错能力。
+CodeActor Agent 协调多个专用智能体——指挥家（Conductor）、仓库分析员（Repo-Analyst）、编码工程师（Coding-Engineer）、对话助手（Chat-Assistant）、运维操作员（DevOps-Operator）和元代理（Meta-Agent）——自主完成复杂的软件工程任务，具备自我纠错能力。
 
 ## 特性
 
 ### 多智能体系统
-- **中枢-辐条架构** — 中央 Conductor 将任务委派给专用子智能体（仓库分析、代码编辑、通用对话）
+- **中枢-辐条架构** — 中央 Conductor 将任务委派给专用子智能体（仓库分析、代码编辑、通用对话、运维操作）
 - **元代理（Meta-Agent）** — 自主设计代理，在运行时为超出内置 Agent 能力的任务动态创建自定义子智能体
 - **自我修正** — `thinking` 工具使 Agent 能够在出错时分析原因并恢复，避免盲目重试
-- **Agent 禁用** — 通过 `--disable-agents=repo,coding,chat,meta` 在启动时有条件地排除子智能体
+- **Agent 禁用** — 通过 `--disable-agents=repo,coding,chat,meta,devops` 在启动时有条件地排除子智能体
 - **ImplPlan 工具** — 状态化实现计划文档，用于复杂多步骤编码任务的分步规划
 
 ### 丰富工具系统（17 个工具）
@@ -42,8 +42,6 @@ CodeActor Agent 协调多个专用智能体——指挥家（Conductor）、仓
 - **调用图分析** — 函数级调用图，含调用者/被调者关系、环路检测、复杂度评分
 - **语义代码搜索** — 通过 LanceDB + SQLite 缓存的向量嵌入（OpenAI `text-embedding-3-small`）
 - **代码骨架/片段** — 批量文件骨架提取和按函数的代码片段检索
-- **文件监听** — 基于 `notify` 的文件系统监听，20s 防抖自动重索引
-- **自动启动** — Go 二进制自动以子进程方式启动 Rust codebase 服务器，包含动态端口分配、健康检查轮询和退出清理
 
 ## 效果截图
 
@@ -90,6 +88,7 @@ enable_streaming = true
 conductor_max_steps = 30
 coding_max_steps = 50
 repo_max_steps = 30
+devops_max_steps = 15
 meta_max_steps = 30
 meta_retry_count = 5
 lang = "Chinese"
@@ -149,10 +148,11 @@ node index.js history                                  # 列出最近任务
 
 | Agent | 工具 | 数量 |
 |-------|-------|-------|
-| Conductor | `delegate_repo`、`delegate_coding`、`delegate_chat`、`delegate_meta`、`finish`、`read_file`、`search_by_regex`、`list_dir`、`print_dir_tree` | 9 |
+| Conductor | `delegate_repo`、`delegate_coding`、`delegate_chat`、`delegate_devops`、`delegate_meta`、`finish`、`read_file`、`search_by_regex`、`list_dir`、`print_dir_tree` | 10 |
 | CodingAgent | 全部 17 个工具（文件、搜索、Shell、thinking、impl_plan、micro_agent） | 17 |
 | RepoAgent | `read_file`、`search_by_regex`、`list_dir`、`print_dir_tree`、`semantic_search`、`query_code_skeleton`、`query_code_snippet` | 7 |
-| ChatAgent | 无（纯 LLM 对话） | 0 |
+| ChatAgent | `micro_agent`、`thinking`、`finish` | 3 |
+| DevOpsAgent | `run_bash`、`read_file`、`list_dir`、`print_dir_tree`、`search_by_regex`、`thinking`、`micro_agent`、`finish` | 8 |
 
 [完整架构文档 →](docs/ARCHITECTURE.md)
 
@@ -184,6 +184,39 @@ meta_retry_count = 5   # JSON 解析失败重试次数（默认 5）
 ./codeactor tui --disable-agents=meta
 ```
 
+## DevOps-Agent（运维代理）
+
+**DevOps-Agent** 是运维和基础设施专家——通过执行 Shell 命令、检查文件系统和分析命令输出来处理所有非编码的运维任务。当 Conductor 遇到系统管理、日志检查、进程管理或 ad-hoc shell 命令类任务时，会通过 `delegate_devops` 委派给 DevOps-Agent。
+
+### 核心能力
+
+- **Shell 命令执行** (`run_bash`) — 运行任意 bash 命令，支持前台/后台运行，含危险检测和工作空间边界检查
+- **文件系统检查** — `read_file`、`list_dir`、`print_dir_tree`、`search_by_regex` 用于浏览日志、配置和目录
+- **自我修正** — 使用 `thinking` 工具分析命令失败原因，调整策略后重试
+- **独立分析** — 使用 `micro_agent` 对命令输出进行深度推理或生成结构化报告
+
+### 示例用例
+
+- 检查磁盘使用率、内存和系统资源
+- 查找最近 24 小时内修改的所有日志文件
+- 重启服务或检查进程状态
+- 检查配置文件
+- 运行系统诊断并生成报告
+- 执行 ad-hoc shell 管道进行数据处理
+
+### 配置
+
+```toml
+[agent]
+devops_max_steps = 15    # DevOps-Agent 最大 LLM 步数（默认 15）
+```
+
+通过启动参数禁用 DevOps-Agent：
+
+```bash
+./codeactor tui --disable-agents=devops
+```
+
 ## Codebase 分析引擎
 
 `codeactor-codebase` 是一个独立的 **Rust** 服务，提供深度代码分析能力。它作为后台 HTTP 服务器运行，由 Go 二进制自动管理。

i18n.go

@@ -44,6 +44,9 @@ type translations struct {
 	HistoryConfirmDelete    string
 	// Confirmation dialog
 	ConfirmDialogHelp string
+	// Command mode
+	CommandModePrompt string
+	CommandModeTips   string
 }
 
 var langMap = map[Language]translations{
@@ -78,6 +81,8 @@ var langMap = map[Language]translations{
 		HistoryKeyClearFilter:     "ctrl+u: 清除过滤",
 		HistoryConfirmDelete:      "确认删除此会话？(y = 确认, 其他键 = 取消)",
 		ConfirmDialogHelp: "←/→ 选择  enter 确认  a 允许  s 全部允许  d/esc 拒绝",
+		CommandModePrompt: "命令模式",
+		CommandModeTips:   "f:下翻页  b:上翻页  i:输入  esc:取消",
 	},
 	LangEnglish: {
 		Title:                            "CodeActor AI Assistant",
@@ -109,7 +114,9 @@ var langMap = map[Language]translations{
 		HistoryKeyBack:            "esc: back",
 		HistoryKeyClearFilter:     "ctrl+u: clear filter",
 		HistoryConfirmDelete:      "Delete this conversation? (y = confirm, any other key = cancel)",
-		ConfirmDialogHelp: "\u2190/\u2192 choose  enter confirm  a allow  s all  d/esc deny",
+		ConfirmDialogHelp: "←/→ choose  enter confirm  a allow  s all  d/esc deny",
+		CommandModePrompt: "COMMAND",
+		CommandModeTips:   "f:pgdn  b:pgup  i:input  esc:cancel",
 	},
 }
 
@@ -190,8 +197,12 @@ func (lm *LanguageManager) GetText(key string) string {
 	case "HistoryConfirmDelete":
 		return translations.HistoryConfirmDelete
 	case "ConfirmDialogHelp":
-		return translations.ConfirmDialogHelp
-	default:
+			return translations.ConfirmDialogHelp
+		case "CommandModePrompt":
+			return translations.CommandModePrompt
+		case "CommandModeTips":
+			return translations.CommandModeTips
+		default:
 		return fmt.Sprintf("[Missing translation: %s]", key)
 	}
 }