メニュー
Comprehensive ReactorPro documentation covering installation, features, agent mode, configuration, skills, MCP servers, and all capabilities. Use this skill when answering questions about ReactorPro features, configuration, troubleshooting, or best practices.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| description | Comprehensive ReactorPro documentation covering installation, features, agent mode, configuration, skills, MCP servers, and all capabilities. Use this skill when answering questions about ReactorPro features, configuration, troubleshooting, or best practices. |
| name | reactorpro-doc |
This skill provides comprehensive documentation for ReactorPro, enabling you to answer questions about its features, configuration, and usage.
ReactorPro supercharges your coding workflow by combining the power of aider with an intuitive desktop interface, enabling seamless AI-assisted code generation, modification, and project management.
.exe installer.dmg and drag ReactorPro to Applications.AppImage and make it executable (chmod +x)When you first open ReactorPro:
~/.reactorproReactorPro offers several chat modes to tailor the AI's behavior to your specific needs:
/code)The default mode for direct interaction with Aider to request code changes. Best for specific, well-defined tasks.
/agent)Activates the autonomous agent. Use for high-level, complex tasks that may require multiple steps, reasoning, and tool use.
/ask)Ask questions about your codebase without the AI attempting to make any changes.
/architect)A specialized mode for planning and executing large-scale changes affecting multiple files. The AI creates a plan first, then uses an "editor" model to apply changes.
/context)Aider will automatically analyze your prompt and try to identify and add the most relevant files to the context.
The Model Library is ReactorPro's advanced provider and model management system.
/add <files>: Adds files to the chat context/drop <files>: Removes files from the context/read-only <files>: Adds files in read-only mode/run <command>: Executes a shell command/test: Runs the predefined test command/clear: Clears the entire chat history and drops all files/clear-logs: Removes only log messages/compact: Summarizes the conversation to reduce token usage/copy-context: Copies context to clipboard in markdown format/tokens: Reports token usage/code: Switch to Code Mode/agent: Switch to Agent Mode/ask: Switch to Ask Mode/architect: Switch to Architect Mode/context: Switch to Context Mode/model <name>: Changes the AI model/reasoning-effort <level>: Sets reasoning effort (low, medium, high)/think-tokens <budget>: Sets thinking token budget/undo: Undoes the last git commit/redo: Re-runs the last prompt/edit-last: Edit and re-submit last message/web <url>: Scrapes URL content and adds to chat/map: Prints the repository map/map-refresh: Forces refresh of repository map/commit: Commits unstaged changes/resolve-conflicts: Resolves merge conflicts using AI/init: Initializes AGENTS.md rule file/handoff <focus>: Extract context and create new focused taskTasks are containers for all the context, conversations, and decisions related to a specific piece of work.
Tasks are stored in .reactorpro/tasks/{taskId}/ with:
settings.json: Task metadata and configurationcontext.json: Messages and files contexttodos.json: Todo items for the taskBreak down complex work into hierarchical structures. Subtasks inherit parent's worktree path.
Git worktrees enable isolated development environments for each task.
Create custom commands in Markdown files with YAML front matter.
~/.reactorpro/commands/.reactorpro/commands/---
description: A brief description of what the command does.
arguments:
- description: Description for the first argument.
required: true
---
Template content using {{1}}, {{2}}, or {{ARGUMENTS}} placeholders.
Lines starting with ! are executed as shell commands.
ReactorPro Memory lets the agent store and retrieve durable, project-scoped knowledge across tasks.
Settings → Memory → EnabledSettings → Agent → Use Memory Toolsretrieve_memory: Search for relevant memoriesstore_memory: Save a new memorylist_memories: List stored memoriesdelete_memory: Delete a specific memorySkills let you extend the agent with reusable, on-demand expertise. A skill is a folder containing instructions (and optionally scripts and reference materials).
~/.reactorpro/skills/.reactorpro/skills/{skills-root}/
{skill-name}/
SKILL.md
references/
scripts/
...
Skills only work when Skills Tools are enabled for the active agent profile:
Settings → Agent → Use Skills ToolsAgentSelector → Use skills toolsHooks allow you to extend ReactorPro's behavior by executing custom JavaScript code.
~/.reactorpro/hooks/.reactorpro/hooks/module.exports = {
onTaskCreated: async (event, context) => {
context.addInfoMessage(`Task "${event.task.name}" was created!`);
},
onPromptStarted: async (event, context) => {
if (event.prompt.includes('secret')) {
return false; // Block the action
}
}
};
onTaskCreated, onTaskInitialized, onTaskClosedonPromptSubmitted, onPromptStarted, onPromptFinishedonAgentStarted, onAgentFinished, onAgentStepFinishedonToolCalled, onToolFinishedonFileAdded, onFileDroppedonCommandExecutedonAiderPromptStarted, onAiderPromptFinishedonQuestionAsked, onQuestionAnsweredonHandleApprovalonSubagentStarted, onSubagentFinishedonResponseMessageProcessedUse the /web command to pull content from any URL into your chat context.
/web https://react.dev/reference/react/useState
.reactorpro/tmp as read-only context fileVoice Control allows speech-to-text functionality to dictate prompts.
Settings → VoiceThe /handoff command extracts relevant context from your current conversation and moves it into a new, focused task.
/handoff [your focus here]
/handoff implement this for teams as well, not just individual users
/handoff execute phase one of the created plan
/handoff check the rest of the codebase and find other places that need this fix
The /compact command summarizes your conversation history to reduce token usage.
Track usage patterns, token consumption, and costs across projects.
ReactorPro includes a built-in MCP server for other MCP-compatible clients.
add_context_file: Add file to contextdrop_context_file: Remove file from contextget_context_files: List files in contextget_addable_files: List available project filesrun_prompt: Execute a promptclear_context: Clear the contextAgent Mode transforms ReactorPro into a powerful, autonomous assistant.
/agent in the prompt fieldAgent Profiles configure the agent's behavior. Create multiple profiles for different workflows.
~/.reactorpro/agents/$projectDir/.reactorpro/agents/profile-name/
├── config.json # Profile configuration
└── rules/ # Additional rule files
├── coding-standards.md
├── architecture.md
└── custom-instructions.md
Subagents are specialized AI assistants that can be delegated specific tasks.
When Agent Mode is active, the agent can interact with Aider:
get_context_files: List files in Aider's contextadd_context_files: Add files to contextdrop_context_files: Remove files from contextrun_prompt: Delegate a coding task to AiderPower tools provide direct file system interaction and development operations.
power---file_read: Read file contentpower---file_write: Create/modify files (overwrite, append, create_only modes)power---file_edit: Targeted find-and-replace changespower---semantic_search: Natural language code searchpower---glob: Find files using glob patternspower---grep: Search content with regexpower---bash: Execute shell commandspower---agent: Delegate to specialized sub-agentsTask tools enable agents to interact with the task management system.
tasks---list_tasks: List all tasks in current projecttasks---get_task: Get comprehensive task detailstasks---get_task_message: Retrieve specific message from task historytasks---create_task: Create new task or subtasktasks---delete_task: Permanently delete a tasktasks---search_task: Semantic search within a tasktasks---search_parent_task: Search parent task content (for subtasks)Extend Agent Mode with external tools through the Model Context Protocol.
{
"mcpServers": {
"server-name": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-puppeteer"]
}
}
}
${projectDir}: Project root directory${taskDir}: Current task's working directoryThe /init command analyzes your codebase and generates an AGENTS.md file.
/init and press EnterThe agent can track and complete multi-step tasks.
set_items: Creates or overwrites the todo listget_items: Reads current todo stateupdate_item_completion: Marks items complete/incompleteclear_items: Clears all itemsAccess settings via the gear icon in the top-right corner.
Manage API keys and connections for LLM providers.
Supported providers:
.env in project directory.aider.conf.yml in project directory.env in home directory.aider.conf.yml in home directoryPass command-line arguments in Settings > Aider > Options.
Set in Settings > Aider > Environment Variables.
.reactorpro/rules/.reactorpro/agents/{profile}/rules/~/.reactorpro/agents/{profile}/rules/ReactorPro uses Handlebars templates for AI agent behavior.
system-prompt.hbs: Main system promptinit-project.hbs: Project initializationworkflow.hbs: Agent workflow guidancecompact-conversation.hbs: Conversation summarizationcommit-message.hbs: Git commit messagestask-name.hbs: Task name generationupdate-task-state.hbs: Task state determinationconflict-resolution-system.hbs: Conflict resolution system promptconflict-resolution.hbs: Conflict resolution instructionshandoff.hbs: Handoff prompt generation$PROJECT_DIR/.reactorpro/prompts/~/.reactorpro/prompts/{{equals value1 value2}}: Check equality{{not value}}: Logical NOT{{assign varName value}}: Assign variable{{increment varName}}: Increment variable{{indent text spaces}}: Indent text{{cdata text}}: Wrap in CDATAMicrophone Not Working (Voice Control)
Power tools not available
Skills not loading
Memory not working
The ultimate network troubleshooting skill — a CCIE/RHCE-level reference and decision engine for diagnosing ANY network, connectivity, bandwidth, latency, DNS, TLS, or security problem. Covers packet-level sensors (NetWatch, RustNet, tshark, ngrep, Pktmon), all capture architectures (SPAN/RSPAN/ERSPAN, transparent Linux bridge, L3 gateway, host-based, one-armed SPAN), Windows AND Linux sensor builds, and Cisco native telemetry (SNMP/NETCONF/gNMI/CLI/pyATS). Includes a decision framework that picks the best tool + architecture per symptom, multi-platform command references, and 20+ worked incident scenarios with implementation details, commands, and scripts. Use whenever something is slow, down, dropping, flapping, unreachable, mis-behaving on the wire, or you need to plan/scale a monitoring sensor fleet.
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
Create AiderDesk agent profiles via interactive Q&A.
Zero-knowledge secrets infrastructure — AI agents manage the complete credential lifecycle without ever seeing values
Comprehensive Agentspan durable workflow orchestration skill for creating, running, scheduling, monitoring, and debugging AI agent workflows using the Agentspan Python SDK and CLI. Covers agents, tools, multi-agent strategies, guardrails, memory, streaming, human-in-the-loop, testing, and production deployment.
Submit software to antivirus vendors to prevent false positive detections. Use when releasing new software that triggers AV warnings, submitting binaries for whitelisting, or checking VirusTotal detection status.