菜单
Control the always-listening Telegram daemon, send messages, manage the task/goal bot, and configure scheduled alerts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | telegram |
| triggers | send telegram, telegram message, telegram listener, start daemon, stop daemon, telegram notification, push notification, telegram bot, @KatanaAgent_bot, telegram alert |
| description | Control the always-listening Telegram daemon, send messages, manage the task/goal bot, and configure scheduled alerts. |
Auto-activate for Telegram message sending, daemon control, or bot configuration.
| What | Command |
|---|---|
| Start listener | /telegram start |
| Stop listener | /telegram stop |
| Status + PID | /telegram status |
| Sleep (alerts only) | /telegram sleep |
| Wake (full response) | /telegram wake |
| View logs | /telegram logs |
/telegram send "Your message here"
/telegram brief # Send morning briefing now
/telegram alert "msg" # System alert notification
Two-tier — daemon is standalone Python, cannot call Claude MCP directly:
| Tier | Mechanism | Commands |
|---|---|---|
| Direct | Pure Python + Supabase + API calls | tasks, goals, habits, weather, oracle vibe, stream status |
| Subprocess | Shell to agent_orchestrator.py | stream send, newsletter gen |
| Queue | Write to /tmp/telegram_agent_queue.json | order, book, vibe activity queries |
telegram_listener.py + telegram_helpers.py~/mcp-servers/telegram-mcp/katana_bot.session/tmp/telegram_listener.pid/tmp/telegram_listener_state.jsonlogs/telegram_listener.log/tmp/telegram_agent_queue.json{ "chat_id": null, "notifications_enabled": true, "morning_brief": true,
"task_reminders": true, "goal_checkin_time": "20:00", "always_listening": true }
→ Full bot command reference: .claude/commands/telegram.md
Extracts and analyzes competitors' ads from ad libraries (Facebook, LinkedIn, etc.) to understand what messaging, problems, and creative approaches are working. Helps inspire and improve your own ad campaigns.
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Install, create, scaffold, and validate new skills in the .claude/skills/ folder. Use when adding a skill from a URL, GitHub repo, zip file, or markdown content, or when creating a new skill from scratch. Also use for validating existing skills, fixing registration issues, or restructuring the .claude folder.
Browser automation via Playwright MCP — reservations, ordering, scraping, form filling. Always confirm before final submission.
Katana setup wizard. Use when user says /setup, 'set up katana', 'configure my agent', 'install features', or 'walk me through setup'. Guides user through API key configuration, skill enabling, and profile creation.
Katana Sensei — life & business orchestrator powered by the 3 I's (Intelligence, Intuition, Initiative). Auto-loads memory on session start (soul.md, user.md, work.md, tasks.md). Runs morning briefings, evening check-ins, task daily iteration, inspirational quotes (Musashi, Stoic, Zen), profile management, and delegates to all agents. Use for any daily routine, task management, memory operations, quote requests, scheduling, or orchestration.