Give your GitHub graph the love your side projects never got.
Keep your graph glowing while you're binging Netflix.
gitfinity is your fully automated GitHub activity generator creating ~3-4 random commits per day so your contribution graph stays alive, consistent, and effortlessly green.
No burnout. No fake projects. Just beautiful green squares.
- Truly random timing: 8 scheduled runs per day with 0-15 minute random delays for unpredictable commit times
- Natural spread: Commits distributed every ~3 hours throughout the entire day (0:00, 3:00, 6:00, 9:00, 12:00, 15:00, 18:00, 21:00 UTC)
- Smart commit distribution: Uses weighted algorithm where most runs create 0 commits, some create 1, rarely 2
- Efficient execution: Short delays (0-15 min) don't waste GitHub Actions time limits
- Realistic patterns: Averages ~3-4 commits per day with natural variation
- Automatic log management: Keeps only the last 100 lines to prevent file bloat
- Fully automated: Uses GitHub Actions (no server required, no costs)
- Simple activity logging: Tracks timestamps in activity.log
The program uses GitHub Actions workflows that run 8 times per day (every ~3 hours):
- 🌙 Midnight: 0:00 UTC (± 0-15 min)
- 🌃 Late night: 3:00 UTC (± 0-15 min)
- 🌅 Early morning: 6:00 UTC (± 0-15 min)
- ☕ Morning: 9:00 UTC (± 0-15 min)
- 🌞 Noon: 12:00 UTC (± 0-15 min)
- 🌤️ Afternoon: 15:00 UTC (± 0-15 min)
- 🌆 Evening: 18:00 UTC (± 0-15 min)
- 🌇 Night: 21:00 UTC (± 0-15 min)
Each run adds a random 0-15 minute delay before executing, adding natural variation without wasting GitHub Actions time!
Each run then executes a Python script that:
- Uses weighted distribution to select 0-2 commits:
- 60% chance: 0 commits (most runs do nothing)
- 35% chance: 1 commit (occasional activity)
- 5% chance: 2 commits (rare bursts)
- For each commit:
- Updates the
activity.logfile with a timestamp - Creates a git commit
- Updates the
- Automatically truncates the log file to 100 lines to prevent bloat
- Pushes all commits to GitHub
Result: ~3-4 commits per day on average, naturally spread throughout 24 hours!
Just a GitHub account! Everything else runs in GitHub Actions (no installation, no server, no costs).
-
Fork this repository
- Click the "Fork" button at the top right of this page
-
Enable the "Daily GitHub Activity" action
- Go to your forked repository
- Click the "Actions" tab
- Enable the "Daily GitHub Activity" workflow
That's it! The workflow will run automatically on schedule.
Edit .github/workflows/daily-commit.yml lines 7-14:
- cron: '0 0 * * *' # 12:00 AM UTC
- cron: '0 3 * * *' # 3:00 AM UTC
- cron: '0 6 * * *' # 6:00 AM UTC
- cron: '0 9 * * *' # 9:00 AM UTC
- cron: '0 12 * * *' # 12:00 PM UTC
- cron: '0 15 * * *' # 3:00 PM UTC
- cron: '0 18 * * *' # 6:00 PM UTC
- cron: '0 21 * * *' # 9:00 PM UTCYou can add/remove runs or change the times. Use crontab.guru for help.
Note: If you reduce the number of scheduled runs, consider adjusting the commit weights in daily_commit.py to maintain desired daily commit counts.
Manual Testing (Optional):
You can test the script locally:
# Run with automatic push
python3 daily_commit.py
# Run without pushing (commits only)
python3 daily_commit.py --no-push
# Test mode: always create 2 commits (useful for testing)
python3 daily_commit.py --test --no-pushEdit .github/workflows/daily-commit.yml line 27:
DELAY=$((RANDOM % 900)) # 900 seconds = 15 minutesChange 900 to adjust the randomization window:
300= 0-5 minutes600= 0-10 minutes900= 0-15 minutes ⭐ Default1800= 0-30 minutes
Edit daily_commit.py around line 33 in the get_s_curve_commits() method:
weights = [60, 35, 5] # 0, 1, 2 commits per runHigher numbers = more likely. Example: [50, 45, 5] makes commits slightly more frequent.
To change the range (currently 0-2), edit line 33:
commits = random.choices(range(3), weights=weights, k=1)[0]Change range(3) to range(4) for 0-3, and update weights accordingly.
Edit daily_commit.py around line 74:
self.truncate_log_file(max_lines=100) # Change max_lines valueEdit daily_commit.py around line 86:
commit_message = f"Daily activity update #{commit_number} - {timestamp}"With the default settings:
Per run (0-2 commits):
- 0 commits: 60% chance (most runs do nothing)
- 1 commit: 35% chance ⭐ Occasional activity
- 2 commits: 5% chance (rare bursts)
Per day (8 runs):
- Expected average: ~3-4 commits per day
- Minimum: 0 commits (all runs skip)
- Maximum: 16 commits (all runs create 2 - very unlikely!)
- Timing: Random within eight 15-minute windows spread across 24 hours
Example day:
0:08 AM → 0 commits (skipped)
3:12 AM → 1 commit
6:02 AM → 0 commits (skipped)
9:14 AM → 0 commits (skipped)
12:05 PM → 1 commit
3:11 PM → 0 commits (skipped)
6:09 PM → 2 commits (rare!)
9:03 PM → 1 commit
Total: 5 commits at unpredictable times!
MIT
This is meant for maintaining GitHub activity and learning automation. The commits are real and will appear in your contribution graph.
Realism features:
- ✅ Commits happen at random times (not predictable patterns)
- ✅ Distributed across 8 time windows throughout the day
- ✅ Variable commit count using weighted distribution
- ✅ Efficient execution (0-15 min delays)
- ✅ Automatic log file management
Contribution graph requirements:
- ✅ Uses your GitHub account email (via noreply address)
- ✅ Commits to default branch (main)
- ✅ Repository must be public
The randomization ensures your activity looks natural and human-like!
