salim 1.12.0

Control your entire laptop from Telegram — screenshots, shell, files, processes, power, browser automation, SSH, vision AI, TTS, scheduler, QR, speed test, file watcher, and more.

🖥️ Salim v11 — AI-Powered Telegram Laptop Control Bot

Control your laptop from anywhere via Telegram. Type in plain English — AI understands and acts.

---

✨ What's New in v11

Feature	Details
🎙️ /mic command	Record from laptop microphone → sends voice note to Telegram
🔊 Real volume control	Uses pulsectl → pactl → amixer (Linux), osascript (Mac)
🚀 Fixed app launcher	Proper setsid + DISPLAY env for GUI apps on Linux
🔍 Smart file search	AI extracts keywords, searches whole laptop, shows buttons
⬇️▶️🗑️📂 File action buttons	Download · Play · Delete · Open for every search result
📄 Real documents	.docx/.xlsx created by AI from natural language
📝 HTML parse mode	All Telegram messages use HTML — no Markdown escaping issues

---

🚀 Quick Start

1. Install

pip install salim
# or from source:
pip install -e .

2. Setup

salim setup

Prompts for:

• Telegram Bot Token — get from @BotFather
• Your Telegram User ID — get from @userinfobot
• AI API key — NVIDIA NIM / Groq / Z.AI (at least one)

3. Run

salim start
salim start --verbose    # with debug logging

---

🤖 AI Natural Language

Just type anything — no need to memorize commands:

"open chrome"
"take a screenshot"
"what's my battery"
"find my Day 03 video"
"turn up the volume"
"mute"
"create a budget spreadsheet for March"
"write me a cover letter for a Python dev role at Google"
"record from microphone for 30 seconds"

---

📋 All Commands

🎙️ Microphone Recording (NEW v11)

/mic              — Record 10s from laptop mic → sends as Telegram voice note
/mic 30           — Record 30 seconds
/mic 10 exec      — Record, transcribe, and execute as AI command
/mic 5 trans      — Record and transcribe only (show text)

Tip: also send any Telegram voice note → transcribed by Whisper → executed as command.

---

🔊 Volume Control

/volume           — Get current volume (e.g. 75%)
/volume 75        — Set to 75%
/volume up        — Increase +10%
/volume down      — Decrease -10%
/volume mute      — Mute audio
/volume unmute    — Unmute audio

Linux: pulsectl → pactl → amixer | Mac: osascript | Windows: PowerShell

---

🚀 Open Apps & URLs

/open chrome           — Launch Google Chrome
/open spotify          — Launch Spotify
/open vscode           — Launch VS Code
/open terminal         — Launch terminal
/open https://yt.com   — Open URL
/open ~/file.pdf       — Open file with default app

Supported: chrome, firefox, safari, brave, vscode, spotify, discord, slack, zoom, telegram, vlc, gimp, obs, terminal, files/nautilus/finder, calculator, settings, libreoffice, word, excel, and any installed app name.

---

🔍 Smart File Search

Type naturally:

"download Day 03"
"find my lecture notes"
"where's the project report"

Results appear as interactive buttons:

1️⃣ 🎬 Day03_Lecture.mp4   ~/Videos/Course   1.2 GB · 3 days ago
   [⬇️ Download]  [🗑️ Delete]  [▶️ Play]

2️⃣ 📝 day03_notes.txt     ~/Documents       45 KB · today
   [⬇️ Download]  [🗑️ Delete]  [👁️ Read]

After download → additional buttons: Play · Delete · Open

---

📄 Document Creation

/doc resume           — AI creates professional resume (.docx)
/doc invoice          — Invoice spreadsheet (.xlsx)
/doc cover letter     — Cover letter (.docx)
/doc budget           — Budget tracker (.xlsx)
/doc meeting minutes  — Meeting minutes (.docx)
/doc report           — Business report (.docx)

Natural language: "write me a resume for Ahmed, software engineer"

Files created locally and sent to Telegram.

---

📸 Screen & Input

/screenshot           — Take screenshot
/ss                   — Same as /screenshot
/stream               — Live screen stream
/streamstop           — Stop stream
/type <text>          — Type on screen
/key ctrl+c           — Press keyboard shortcut
/click <x> <y>        — Mouse click
/scroll <n>           — Scroll (+ up, - down)
/move <x> <y>         — Move mouse cursor
/mousepos             — Get current mouse position
/brightness <0-100>   — Set screen brightness
/notify <title>|<msg> — Desktop notification

---

💻 System & Shell

/info     — Full system overview (CPU, RAM, disk, network)
/cpu      — CPU usage & cores
/mem      — Memory usage
/disk     — Disk usage
/battery  — Battery % and charging status
/network  — Network interfaces and IPs
/uptime   — System uptime
/top      — Live process monitor
/ps [name]— Running processes
/kill <pid>— Kill process
/run <cmd>— Run shell command
/runbg <cmd>— Run in background
/env      — Environment variables
/which <prog>— Find binary path
/cron <cmd> <schedule>— Add cron job
/cronjobs — List cron jobs
/cronstop <id>— Remove cron job

---

📂 Files

/ls [path]    /cd <path>    /pwd
/cat <file>   /stat <file>  /find <pattern>
/grep <pattern> [path]      /mkdir <name>
/rm <path>    /mv <s> <d>   /cp <s> <d>
/write <file> <text>
/download <path>   — Send file to Telegram
/upload            — Receive file from Telegram
/zip <path>        /unzip <file>

---

📋 Clipboard

/copy <text>    — Copy text to clipboard
/paste          — Read clipboard (text or image)
/copyphoto      — Copy photo to clipboard (reply to photo)

---

🌐 Browser Automation

/browse <url>             — Open URL + screenshot
/browse_click <selector>  — Click element
/browse_type <sel> <text> — Type in field
/browse_scroll [up|down]  — Scroll page
/browse_js <javascript>   — Run JS
/browse_back              — Go back
/browse_close             — Close browser

---

🗣️ Text-to-Speech

/tts <text>         — Laptop speaks text
/say <text>         — Same as /tts
/tts_lang <code>    — Language (en, ar, fr...)
/tts_speed <0.5-2>  — Speed
/tts_engine         — Switch engine
/tts_voice          — List voices

---

🔌 SSH

/ssh_add <alias> <user@host> [port]
/ssh <alias> <command>
/ssh_list   /ssh_remove <alias>
/ssh_upload <alias> <local> <remote>
/ssh_download <alias> <remote>

---

👁️ Vision AI

/vision on      — Auto-describe all photos
/askvision <q>  — Ask about last photo

Send any photo → reply with a question.

---

🔋 Power

/shutdown   /restart   /sleep   /hibernate
/lock       /logout    /screensaver

Destructive commands ask for confirmation.

---

📷 Camera & Recording

/cam           — Webcam snapshot
/cam list      — List cameras
/record <sec>  — Record screen video

---

🔔 Alerts

/alert add cpu > 80       — Alert when CPU > 80%
/alert add battery < 20   — Low battery alert
/alert list               — View alerts
/alert del <id>           — Remove alert

---

📝 Notes

/note <text>   — Save note
/notes         — List notes
/notedel <id>  — Delete note
/noteclear     — Clear all

---

⏰ Scheduler

/at 14:30 screenshot    — Run at specific time
/every 5m screenshot    — Run on repeat
/jobs                   — List jobs
/jobcancel <id>         — Cancel job

---

📡 File Watcher

/watch <path>    — Watch for changes
/watchlist       — List watchers
/watchstop <id>  — Stop watcher

---

⚡ Other

/speedtest          — Internet speed (download/upload/ping)
/qr <text>          — Generate QR code
/qrread             — Read QR from photo
/code <desc>        — AI writes & runs Python code
/guard on/off       — Intrusion detection (webcam)
/config             — Bot settings
/logs [n]           — Audit log
/history            — Conversation memory
/uploads            — Upload history

---

🔧 System Dependencies

Linux:

sudo apt install -y ffmpeg portaudio19-dev pulseaudio-utils
pip install pulsectl sounddevice scipy

Mac:

brew install ffmpeg portaudio
pip install sounddevice scipy

Windows: install ffmpeg from ffmpeg.org, add to PATH.

---

🔑 AI Providers (at least one required)

Provider	Free Tier	Setup
NVIDIA NIM	Credits	build.nvidia.com
Groq	Yes	console.groq.com
Z.AI	Yes (FREE)	z.ai

---

🔒 Security

• Only your Telegram user ID can control the bot
• All commands logged (/logs)
• Destructive commands require confirmation
• Intrusion guard alerts via Telegram

---

Salim v11 — Python · python-telegram-bot · faster-whisper · sounddevice · pulsectl