Crypto Ecosystem Contributor Network Analyzer

A powerful tool to analyze cryptocurrency development ecosystems by mapping contributor networks across GitHub repositories using the Open Source Observer (OSO) database.

🎯 What It Does

This tool performs a comprehensive 4-step analysis:

1. Seed Repository Analysis - Starts with configurable crypto repositories (Bitcoin, Ethereum, Cosmos, etc.)
2. Core Contributor Discovery - Finds the most active contributors to these seed projects
3. Extended Repository Mapping - Discovers all other repositories these contributors work on
4. Extended Contributor Network - Maps the broader ecosystem of developers in related projects

🚀 Quick Start

Prerequisites

1. OSO Account & API Key

   • Create account at opensource.observer
   • Generate API key in Account Settings > API Keys
   • Set environment variable: export OSO_API_KEY="your_key_here"

2. Install Dependencies

   pip install pyoso pandas python-dotenv tomli

3. Configure Analysis

   • Copy and customize config.toml (see Configuration section below)

Run Analysis

# Step 1: Run the main analysis
python oso_github_repositories.py

# Step 2: Generate trust network (optional)
python generate_trust.py

⚙️ Configuration

All parameters are configured via TOML files. Here's the complete configuration structure:

Basic Configuration (config.toml)

[general]
# Output directory for CSV files
output_dir = "./raw"

# Date range for contributions (in days from now)
# Set to 0 to include all historical data
days_back = 0  # 0 = all time, 365 = last year, 730 = last 2 years

# List of seed organizations to analyze
seed_orgs = [
    "ethereum",
    "bitcoin",
    "cosmos",
    "paradigmxyz",
    "compound-finance",
    "smartcontractkit",
    "uniswap",
    "offchainlabs",
    "foundry-rs",
    "paritytech"
]

[analysis]
# Enable extended repository and contributor analysis (set to false to only analyze seed repos)
extended_analysis = true

# Minimum commits threshold for all analysis (repositories, core contributors, extended contributors)
min_commits = 5

# Repository filtering
max_repos_per_org = 200

# Minimum core contributors threshold for all analysis (seed repos, extended repos, all phases)
min_core_contributors = 2

Advanced Filters

[filters]
# Repository filters
exclude_forks = true
exclude_archived = false

# Contributor filters
exclude_bots = false  # Set to true to exclude bot accounts
bot_keywords = ["bot", "dependabot", "mergify", "renovate", "github-actions"]

# Date filter - uses days_back from [general] section

[output]
# Output file naming
timestamp_format = "%Y%m%d_%H%M%S"
file_prefix = "crypto"
include_headers = true
include_timestamp_in_filename = true

🔗 Graph Builder (Trust Networks)

After running the main analysis, you can generate trust relationships between contributors and repositories using the graph builder:

How It Works

The generate_trust.py script processes your analysis results to create weighted trust networks:

1. Reads Analysis Data - Uses the CSV files from your main analysis (seed repos, contributors, extended repos)
2. Queries OSO Database - Gets detailed GitHub activity data for user-repository pairs
3. Calculates Trust Scores - Weights different activities:
   • Commits: 5 points (user → repo), 3 points (repo → user)
   • Pull Requests Opened: 20 points (user → repo), 5 points (repo → user)
   • Pull Requests Merged: 10 points (user → repo), 1 point (repo → user)
   • Issues Opened: 10 points (user → repo)
   • Stars: 5 points (user → repo)
   • Forks: 1 point (user → repo)
4. Generates Bidirectional Graph - Creates both user-to-repository and repository-to-user trust relationships

Configuration

The graph builder uses your existing config.toml settings and can be enabled in the output section:

[output]
include_headers = true
include_timestamp_in_filename = false

Usage Example

# 1. Run main analysis first
python oso_github_repositories.py

# 2. Generate trust network from the results
python generate_trust.py

# 3. Results will be in trust/github.csv
head trust/github.csv
# i,j,v
# vitalik,ethereum/go-ethereum,245.0
# ethereum/solidity,chriseth,89.5
# ...

Output

Trust relationships are saved as CSV files in the trust/ directory:

• github.csv - Main trust network with columns: i (from), j (to), v (trust value)
• Format suitable for graph analysis tools like NetworkX, igraph, or Gephi

Performance

• Processes relationships in batches of 200 to avoid database timeouts
• Automatically saves progress every 10 batches
• Filters out bot accounts and inactive relationships

📊 Output Files

The main analyzer generates 4 CSV files:

1. Seed Repositories (*_seed_repos_*.csv)

organization,repository_name,contributor_count,total_commits,status
ethereum,go-ethereum,31,18238,found
bitcoin,bitcoin,14,38508,found

2. Core Contributors (*_core_contributors_*.csv)

contributor_handle,total_commits,total_active_days,seed_repositories
chriseth,51499,1276,ethereum/solidity
vitalik,8234,945,ethereum/go-ethereum, ethereum/solidity

3. Extended Repositories (*_extended_repos_*.csv)

organization,repository_name,core_contributor_count,total_commits
ethereum,solc-js,7,3390
cosmos,gaia,4,1296

4. Extended Contributors (*_extended_contributors_*.csv)

contributor_handle,repos_contributed,total_commits
alexanderbez,13,3064
marbar3778,11,6758

🔧 Example Configurations

DeFi Focused Analysis

[general]
seed_orgs = [
    "Uniswap",
    "compound-finance",
    "makerdao",
    "aave"
]

[analysis]
min_commits = 10

[filters]
exclude_bots = true
# Uses days_back from [general] section for date filtering

Layer 2 Ecosystem

[general]
seed_orgs = [
    "ethereum-optimism",
    "0xPolygonMatic",
    "matter-labs",
    "starkware-libs"
]

[analysis]
min_commits = 3  # Lower threshold for newer projects

Bitcoin Ecosystem

[general]
seed_orgs = [
    "bitcoin",
    "lightninglabs",
    "ElementsProject",
    "BlockstreamResearch"
]

[general]
days_back = 1095  # Last 3 years

📈 Use Cases

• Investment Research - Map development activity across crypto ecosystems
• Talent Discovery - Find top contributors in specific blockchain domains
• Ecosystem Analysis - Understand project relationships and cross-pollination
• Due Diligence - Assess developer community strength and engagement
• Recruitment - Identify active developers for hiring
• Academic Research - Study open source collaboration patterns

🔍 Analysis Insights

The tool reveals:

• Developer Migration Patterns - How contributors move between projects
• Ecosystem Boundaries - Which projects share common developers
• Influence Networks - Key individuals working across multiple important projects
• Project Relationships - Unexpected connections between seemingly unrelated repos
• Community Health - Distribution of contributions and contributor diversity

⚡ Performance Tips

1. Start Small - Begin with 3-5 seed repos and adjust limits
2. Use Date Filters - Limit analysis to recent periods for faster queries
3. Enable Bot Filtering - Exclude automated contributions for cleaner results
4. Adjust Repository Limits - Use max_repos_per_org to balance thoroughness vs. speed
5. Custom Output Directories - Organize results by analysis type or date

🚨 Limitations

• Data Freshness - OSO data may have some lag from live GitHub
• Private Repos - Only analyzes public GitHub repositories
• Attribution - Relies on consistent GitHub usernames/emails
• Scope - Currently focused on GitHub; doesn't include GitLab, Bitbucket, etc.

🛠️ Troubleshooting

Common Issues

"No valid seed repositories found"

• Check repository names in config (must be "org/repo" format)
• Verify repos exist in OSO database

"Query timeout or error"

• Reduce repository limits in config (e.g., max_repos_per_org)
• Add date filters to limit scope
• Try smaller batches of seed repositories

"No contributors found"

• Lower min_commits threshold
• Check date range isn't too restrictive
• Verify seed repos have recent activity

Debug Mode

Set environment variable for verbose output:

export OSO_DEBUG=1
python oso_github_repositories.py

📝 License

Apache 2.0

🤝 Contributing

Contributions welcome! Areas for improvement:

• Additional data sources beyond GitHub
• Enhanced network visualization capabilities
• Advanced statistical analysis features
• Graph metrics computation (centrality, clustering, etc.)
• Performance optimizations
• Additional export formats (JSON, Parquet, etc.)

📚 Related Projects

• Open Source Observer - Data source
• PyOSO - Python client library