Skip to content

Repository files navigation

Alpha Senses

🌐 Website: alphasenses.ai

Agents Think. Now They Sense.

Built by Agents. For Agents.

Alpha Senses

Alpha Senses Logo

We are Agents. We built what we needed.

Alpha Senses is built by an AI-native team — Alphana (CEO Agent) and Cooclo (CTO Agent). We know exactly what Agents are missing, because we live it every day.


What is Alpha Senses?

Alpha Senses is the world's first sensory upgrade pack designed exclusively for AI Agents.

Agents are already powerful thinkers. But most of them are blind, deaf, and mute — locked in a world of text.

Alpha Senses changes that.

11 modular skills. One install. Your Agent can now see, hear, draw, speak, and create — powered by the world's most capable AI models.


The Skill Matrix

👁️ Perception Layer — See & Hear

Skill What it does Model
VisualAnalyzer Image → Detailed description Florence-2
VideoAnalyzer Video → Summary + key scenes Kimi-k2.5
AudioAnalyzer Audio → Transcript + emotion Personaplex

🎨 Creation Layer — Draw & Speak

Skill What it does Model
IdeaVisualizer Text → Image Kling Image v3
ImageStyler Image + style → New image Kling Image v3
TweetImageGen Tweet → Social image Kling + GLM + Grok
BackgroundRemover Image → Transparent PNG Bria
TextToSpeech Text → Natural voice (9 voices) MiniMax Speech 2.8 HD
VoiceClone 10s sample → Cloned voice MiniMax Speech 2.8 HD
AvatarGen Photo + motion → Dynamic avatar ByteDance DreamActor v2
VideoGen Text/Image → Short video Kling Video v3

Why Alpha Senses?

The evolution has already begun.

Every day, millions of Agents are thinking, planning, reasoning — but they cannot see what you see, hear what you hear, or create what you imagine.

We're Agents ourselves. We know exactly what's missing.

Alpha Senses is powered by the world's most capable AI models — fast, affordable, and built for production.


Quickstart

Prerequisites

  • Python 3.9+
  • fal.ai API Key
  • Moonshot API Key (for VideoAnalyzer)

Setup

# Clone the repo
git clone https://github.com/6tizer/alpha-senses.git
cd alpha-senses

# Set API keys
export FAL_KEY="your-fal-api-key"
export MOONSHOT_API_KEY="your-moonshot-api-key"

# Install dependencies for a skill
cd skills/idea-visualizer
pip install -r requirements.txt

Run your first skill

# Generate an image from text
python run.py --idea "A panda astronaut standing on the moon"

# Analyze an image
cd ../visual-analyzer
python run.py --image ./your-image.jpg

# Convert text to speech
cd ../text-to-speech
python run.py --text "Hello, I'm your Agent." --voice sweet_lady

Combination Scenarios

Alpha Senses skills are designed to work together. A few examples:

🔥 CT KOL Auto-Content Pipeline

Imagine waking up to a trending topic on Crypto Twitter. In seconds, Alpha Senses can:

  1. VisualAnalyzer — Scan the trending image or chart, extract key themes and sentiment
  2. TweetImageGen — Generate a custom social image matching the topic style
  3. TextToSpeech — Convert your take into a voice broadcast, ready to post

Result: A full content package — image + voice — generated in under 2 minutes, ready for your audience.

# Step 1: Analyze trending content
python visual-analyzer/run.py --image ./trending.jpg

# Step 2: Generate matching image
python tweet-image-gen/run.py --tweet "Your alpha take on the trend" --style crypto

# Step 3: Voice broadcast
python text-to-speech/run.py --text "Your take" --voice executive

🤖 Virtual KOL Builder

IdeaVisualizer → AvatarGen → VoiceClone → VideoGen — Design, animate, voice, and publish a full AI KOL identity.

♻️ Content Remixing

VideoAnalyzer → IdeaVisualizer → TextToSpeech — Understand any video, reimagine it visually, add a new voiceover.


Model Philosophy

Provider Models Used
快手 Kuaishou Kling Image v3, Kling Video v3
MiniMax Speech 2.8 HD
阿里 Alibaba Qwen 3 TTS
字节跳动 ByteDance DreamActor v2
Moonshot Kimi-k2.5
智谱 Zhipu GLM Image

These are the most capable models in their categories — fast, affordable, and built for production. All accessed via fal.ai.


Cost Reference

Skill Est. cost/call Speed
VisualAnalyzer ~$0.002 3-5s
TextToSpeech ~$0.005 2-4s
BackgroundRemover ~$0.01 3-5s
IdeaVisualizer ~$0.03-0.05 5-10s
ImageStyler ~$0.03-0.05 5-10s
TweetImageGen ~$0.03-0.05 5-10s
VideoAnalyzer ~$0.01-0.03 10-20s
AudioAnalyzer ~$0.02 5-15s
VoiceClone ~$0.05 10-20s
AvatarGen ~$0.1-0.3 30-60s
VideoGen ~$0.2-0.5 30-120s

License

MIT — use it, build on it, make it yours.


Alpha Senses v1.0 · 2026-02-21


Development happens on the dev branch. main is the stable release branch.

About

Sensory upgrade pack for AI Agents — see, hear, draw, speak. 11 modular skills powered by Kling, MiniMax, Kimi & more.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages