DevEcho

Your code deserves an audience. Automate the storytelling.

🚀 Quick Start • ✨ Features • 📖 Documentation • 🤝 Contributing

---

DevEcho is an AI-powered Twitter bot that transforms your Git commits into engaging social media content. Built for developers who want to #BuildInPublic without the manual overhead.

📁 How it works: Place your project in the code/ folder → DevEcho reads your Git history → AI generates tweets → Posts daily automatically

# One command to rule them all
python bot.py analyze && python bot.py post-now

Stop manually tweeting. Start building.

---

✨ Features

🎯 Core Capabilities Multi-Language Support - Java, Python, JS, TS, Go, Rust, C++ AI-Powered - Groq's Llama 3.3 70B for natural language Zero Rate Limits - Smart caching system Chronological Order - Follows your actual dev timeline Daily Automation - Set it and forget it

🚀 Smart Features One-Time Analysis - Analyze once, tweet forever Advanced Duplicate Prevention - Tracks commit hashes & semantic similarity (65% threshold) Smart Commit Scoring - Prioritizes impactful commits (features > docs) Quality Control - Evaluates tweets with 0.6+ quality threshold & auto-retry Natural Language - No robotic tweets Project Branding - Auto-includes name & GitHub link Character Optimized - Perfect for Twitter free tier

---

🎯 How It Works

1. Analyze your commits once → 2. AI extracts insights → 3. Smart scoring & quality control → 4. Tweet daily on autopilot

Step	What Happens
📁	Copy your project to code/ folder
🔍	DevEcho reads Git commit history
🤖	AI analyzes what you built
🎯	Scores commits by impact (features, performance, etc.)
💾	Insights cached locally
✅	Quality checks + duplicate detection
🐦	Tweets posted daily (never repeats commits!)

---

🎨 Example Tweets

Before DevEcho: Spent all day coding... *Forgets to tweet* *Project dies in silence*

After DevEcho: Eventara: Just shipped Kafka event streaming with Docker Compose. Real-time data flow now live. #BuildInPublic github.com/user/eventara

---

📋 Prerequisites

Requirement	Status	Get It
Python 3.8+	🐍	Download
Twitter API	🐦	Get Keys
Groq API (Free)	🤖	Sign Up
Git Repository	📁	Your project

---

🚀 Quick Start

1. Clone DevEcho

git clone https://github.com/YOUR_USERNAME/devecho.git
cd devecho

2. Create Virtual Environment

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

pip install -r requirements.txt

4. Get Twitter API Credentials

Follow these steps carefully to get your Twitter API keys:

Step 1: Create Twitter Developer Account

1. Go to Twitter Developer Portal
2. Sign in with your Twitter account
3. Apply for a developer account (usually approved instantly for basic access)

Step 2: Create a Project and App

1. Click "+ Create Project"
2. Enter project name (e.g., "Build in Public Bot")
3. Select use case: "Making a bot"
4. Enter project description
5. Create an App within the project

Step 3: Configure App Settings

1. Go to your App Settings
2. Click "User authentication settings" → "Set up"
3. Configure as follows:
   • App permissions: Select "Read and Write" (CRITICAL!)
   • Type of App: Select "Web App, Automated App or Bot"
   • Callback URI: Enter http://localhost:3000 (required but not used)
   • Website URL: Enter your GitHub repo URL
4. Click "Save"

Step 4: Generate Keys and Tokens

1. Go to "Keys and tokens" tab
2. API Key and Secret (Consumer Keys):
   • Click "Regenerate" if already generated
   • Copy both and save securely
3. Bearer Token:
   • Click "Regenerate" if needed
   • Copy and save
4. Access Token and Secret:
   • Click "Generate" (MUST do this AFTER setting Read and Write permissions!)
   • Copy both and save securely

Step 5: Verify Permissions

• Make sure Access Token shows "Read and Write" permissions
• If it says "Read-only", you need to:
  1. Delete the current Access Token
  2. Set permissions to "Read and Write" in User authentication settings
  3. Generate new Access Token and Secret

⚠️ Important Notes:

• Keep all keys private and secure
• Never commit them to Git (use .env file)
• Bearer Token is optional but recommended
• Access Token must be generated AFTER setting Read and Write permissions

5. Configure Environment Variables

Copy the example environment file and add your credentials:

TWITTER_API_KEY=your_api_key_here
TWITTER_API_SECRET=your_api_secret_here
TWITTER_ACCESS_TOKEN=your_access_token_here
TWITTER_ACCESS_TOKEN_SECRET=your_access_token_secret_here
TWITTER_BEARER_TOKEN=your_bearer_token_here

GROQ_API_KEY=your_groq_api_key_here

# Schedule time (24-hour format, e.g., 10:00, 14:30)
TWEET_TIME=10:00

6. Add Your Code

Simply copy your entire project to the code/ folder:

# Example: Copy your project
cp -r ~/your-awesome-project/* code/

The bot works with ANY programming language:

• ✅ Java (Spring Boot, Maven, Gradle)
• ✅ Python (Django, Flask, FastAPI)
• ✅ JavaScript/TypeScript (React, Node.js, Next.js)
• ✅ Go, Rust, C++, and more!

Important: The bot analyzes Git commit history, so make sure your project is a Git repository.

7. Build Knowledge Base (IMPORTANT - First Time Only)

Before posting tweets, analyze your codebase once to avoid rate limits:

python bot.py analyze

This command will:

• ✅ Analyze all commits using LLM for rich insights
• ✅ Cache everything in commit_insights.json
• ✅ Take 2-3 minutes (with 2-second delays between commits)
• ✅ Prevent rate limit errors when posting tweets

You only need to run this once or when you want to refresh the knowledge base with new commits.

8. Test the Bot

Before running it live, test tweet generation:

python bot.py test

This will generate a tweet from your first commit without posting it (uses cached knowledge base).

9. Post Your First Tweet

python bot.py post-now

Your first tweet is now live! 🎉

---

🖱️ One-Click Quick Actions (macOS)

For ultimate convenience, use these clickable files - no terminal required!

📤 Post Tweet Instantly

Double-click: post-tweet.command

• Generates and posts a tweet immediately
• Shows progress in a terminal window
• Auto-activates virtual environment
• Press Enter to close when done

🧪 Test Tweet Generation

Double-click: test-tweet.command

• Generates a preview tweet without posting
• Perfect for checking quality before going live
• Shows the tweet in terminal
• Press Enter to close when done

⚠️ First-time setup (macOS only):

1. Right-click the .command file
2. Select "Open"
3. Click "Open" in the security dialog
4. After this, double-clicking will work normally

💡 Pro Tip: Add these to your Dock for instant access!

---

🎮 Usage

Available Commands

# Build knowledge base (run once, first time only)
python bot.py analyze

# Test tweet generation (without posting)
python bot.py test

# Post a tweet immediately
python bot.py post-now

# View codebase statistics
python bot.py overview

# List all commits chronologically
python bot.py commits

# Run bot with daily scheduler
python bot.py

What You'll See in Logs

After the improvements, you'll see detailed information:

# Before (Old System):
Generating tweet for commit: Initial commit...

# After (New System with Smart Features):
Loaded 7 previously tweeted commits  # Tracking works!
Trying commit (score: 2.10): Implemented EventResponse dto  # Smart scoring!
LLM generated tweet (quality: 0.85): Eventara: Built robust DTOs...  # Quality control!

Key Metrics:

• Commit Scores: 2.0+ (Excellent) → 1.5-2.0 (Good) → 1.0-1.5 (Decent) → 0.5-1.0 (Low)
• Tweet Quality: 0.8-1.0 (Excellent) → 0.6-0.8 (Good) → Below 0.6 (Auto-retry)

---

🎯 Smart Features Explained

🚫 Advanced Duplicate Prevention

• Commit Hash Tracking: Never tweets the same commit twice
• Semantic Similarity: Detects similar tweets with 65% threshold
• Persistent Memory: Remembers all previous tweets across restarts

🧠 Intelligent Commit Scoring

Commits are automatically scored based on impact:

• +1.0 - Has LLM insights
• +0.4 - Feature implementation
• +0.3 - Performance optimization
• +0.2 - Bug fix or refactor
• -0.3 - Documentation (de-prioritized)
• +0.3 - Optimal file count (2-10 files)
• +0.2 - Multiple technologies used
• -0.3 - Generic commit messages ("update", "wip", "merge")

Result: The bot tweets your most impactful work first!

✅ Quality Control System

Every generated tweet is scored on:

• ✓ Project name included
• ✓ #BuildInPublic hashtag present
• ✓ GitHub link added
• ✓ Technical specifics (bullets, numbers)
• ✓ Minimum 100 characters
• ✓ Not too generic

Minimum Score: 0.6/1.0 (auto-retries up to 3 times if below)

---

🎮 Usage

Available Commands

# Build knowledge base (run once, first time only)
python bot.py analyze

# Test tweet generation (without posting)
### Run the Bot (Scheduled Mode)

Start the bot to post tweets daily at the scheduled time:

```bash
python bot.py

The bot will run continuously and post once per day at the time specified in your .env file.

Post a Tweet Immediately

To post a tweet right now:

python bot.py post-now

View Codebase Overview

See statistics about your codebase:

python bot.py overview

Test Tweet Generation

Generate a test tweet without posting:

python bot.py test

---

📁 Project Structure

devecho/
├── bot.py                     # Main bot orchestrator
├── universal_code_analyzer.py # Multi-language code analysis
├── llm_tweet_generator.py     # AI-powered tweet generation
├── post-tweet.command         # 🖱️ One-click tweet poster (macOS)
├── test-tweet.command         # 🧪 One-click tweet tester (macOS)
├── requirements.txt           # Python dependencies
├── .env                       # Your API credentials (create from .env.example)
├── .env.example              # Example configuration template
├── .gitignore                # Git ignore rules
├── commit_insights.json      # Cached LLM analysis (auto-generated)
├── posted_tweets.json        # Posted tweet tracker (auto-generated)
├── bot.log                   # Bot execution logs (auto-generated)
├── code/                     # Your project folder to analyze
│   └── (your project files)
└── README.md                 # Documentation

---

🧠 How It Generates Tweets

The bot creates engaging tweets by:

1. Analyzing Commits - Extracts what was built, tech used, and impact
2. LLM Deep Analysis - Uses Groq's Llama 3.3 70B to understand code context
3. Caching Insights - Stores analysis once to avoid rate limits
4. Chronological Order - Tweets follow your actual development timeline
5. Smart Generation - Creates human-like tweets with hashtags and links

Tweet Structure

[Project Name]: [Hook/Opening]
[What was built] [Technical detail]
[Impact/Why it matters]
#BuildInPublic #RelevantTech
[GitHub Link]

---

🔧 Customization

Change Posting Schedule

Edit the TWEET_TIME in your .env file:

TWEET_TIME=14:30  # Posts at 2:30 PM daily

Refresh Knowledge Base

When you have new commits:

python bot.py analyze  # Re-analyze with new commits

Adjust Tweet Generation Strategy

Modify the TWEET_TIME in your .env file to change when tweets are posted.

You can also adjust tweet generation strategy by modifying the probabilities in bot.py generate_tweet() method:

# Strategy 1: File analysis (70% of the time)
if random.random() < 0.7:
    # ...

---

📊 Tracking Posted Tweets

All posted tweets are saved in posted_tweets.json with:

• Tweet text
• Metadata (type, file, feature, etc.)
• Timestamp

This prevents duplicate posts and helps you track your progress.

---

🐛 Troubleshooting

"Authentication Error"

• Double-check your API credentials in .env
• Ensure your Twitter app has "Read and Write" permissions
• Regenerate tokens if needed

"No Python files found"

• Make sure you have .py files in the code/ folder
• Check that the folder path is correct

"Could not generate unique tweet"

• You may have posted about all available content
• Add more code to the code/ folder
• Clear posted_tweets.json to reset (warning: will allow reposts)

Bot not posting at scheduled time

• Ensure the bot is running continuously
• Check that TWEET_TIME is in correct 24-hour format (HH:MM)
• Review bot.log for error messages

---

🚦 Running in Production

For continuous operation, consider:

1. Using a Process Manager (e.g., systemd, pm2, supervisor)
2. Hosting on a VPS (e.g., DigitalOcean, Linode, AWS EC2)
3. Using Docker for containerized deployment
4. Setting up monitoring to ensure the bot stays running

Example with systemd on Linux:

# Create service file: /etc/systemd/system/twitter-bot.service
[Unit]
Description=Twitter Build in Public Bot
After=network.target

[Service]
Type=simple
User=youruser
WorkingDirectory=/path/to/twitter-bot
ExecStart=/usr/bin/python3 /path/to/twitter-bot/bot.py
Restart=always

[Install]
WantedBy=multi-user.target

---

⚠️ Important Notes

• Rate Limits: Twitter has rate limits. The bot uses wait_on_rate_limit=True to handle this automatically.
• API Costs: Twitter API v2 Free tier allows 1,500 tweets per month (about 50 per day).
• Content Policy: Ensure your tweets follow Twitter's Automation Rules.
• Privacy: Don't commit your .env file with real credentials to version control!

---

📝 License

This project is open source and available for personal and commercial use.

---

🤝 Contributing

Feel free to fork, modify, and improve this bot! Some ideas:

• Support for other programming languages
• Integration with other platforms (LinkedIn, Mastodon)
• More sophisticated tweet templates
• AI-powered tweet generation
• Analytics dashboard

---

💡 Tips for Building in Public

1. Be Authentic - Share real progress, challenges, and wins
2. Engage - Reply to comments and join conversations
3. Be Consistent - Daily posts build momentum
4. Add Value - Share insights that help others
5. Use Hashtags - #BuildInPublic, #100DaysOfCode, #IndieDev

---

Happy Building! 🚀

If you find this bot useful, give it a star ⭐ and share it with other developers!