Skip to main content

Upload Framework Packages to GitHub Releases

Prerequisites​

  1. GitHub Personal Access Token:

    # Set environment variable
    export GITHUB_TOKEN=your_token_here

    # Or pass via --token flag

  2. Git Tags Created:

    cd /Users/rms/Documents/projects/ai-dev-kit

    # Create tags for each framework
    git tag -a kanban-v2.1.0 -m "Release kanban framework v2.1.0"
    git tag -a workflow-mgt-v2.1.4 -m "Release workflow mgt framework v2.1.3"
    git tag -a numbering-versioning-v2.0.0 -m "Release numbering & versioning framework v2.0.0"
    git tag -a doc-lifecycle-v1.0.0 -m "Release doc-lifecycle framework v1.0.0"
    git tag -a debug-path-v1.0.0 -m "Release debug-path framework v1.0.0"

    # Push tags to remote
    git push origin --tags

  3. Packages Built:

    • Packages must exist in packages/frameworks/dist/packages/
    • Run build script first if not already built
cd /Users/rms/Documents/projects/ai-dev-kit

# Set GitHub token
export GITHUB_TOKEN=your_token_here

# Upload all packages
python3 "packages/frameworks/workflow mgt/scripts/upload_all_packages.py" \
    --token $GITHUB_TOKEN \
    --repo earlution/ai-dev-kit \
    --verbose

Dry Run (Test First)​

python3 "packages/frameworks/workflow mgt/scripts/upload_all_packages.py" \
    --token $GITHUB_TOKEN \
    --dry-run

Method 2: Upload Individual Packages​

cd /Users/rms/Documents/projects/ai-dev-kit
PACKAGES_DIR="packages/frameworks/dist/packages"
UPLOAD_SCRIPT="packages/frameworks/workflow mgt/scripts/upload_to_github_release.py"
REPO="earlution/ai-dev-kit"

# Upload kanban package
python3 "$UPLOAD_SCRIPT" \
    "$PACKAGES_DIR/kanban-v2.1.0.tar.gz" \
    --hash-file "$PACKAGES_DIR/kanban-v2.1.0.tar.gz.sha256" \
    --tag kanban-v2.1.0 \
    --token $GITHUB_TOKEN \
    --repo $REPO \
    --verbose

# Upload workflow mgt package
python3 "$UPLOAD_SCRIPT" \
    "$PACKAGES_DIR/workflow-mgt-v2.1.4.tar.gz" \
    --hash-file "$PACKAGES_DIR/workflow-mgt-v2.1.4.tar.gz.sha256" \
    --tag workflow-mgt-v2.1.4 \
    --token $GITHUB_TOKEN \
    --repo $REPO \
    --verbose

# Upload numbering & versioning package
python3 "$UPLOAD_SCRIPT" \
    "$PACKAGES_DIR/numbering-versioning-v2.0.0.tar.gz" \
    --hash-file "$PACKAGES_DIR/numbering-versioning-v2.0.0.tar.gz.sha256" \
    --tag numbering-versioning-v2.0.0 \
    --token $GITHUB_TOKEN \
    --repo $REPO \
    --verbose

# Upload doc-lifecycle package
python3 "$UPLOAD_SCRIPT" \
    "$PACKAGES_DIR/doc-lifecycle-v1.0.0.tar.gz" \
    --hash-file "$PACKAGES_DIR/doc-lifecycle-v1.0.0.tar.gz.sha256" \
    --tag doc-lifecycle-v1.0.0 \
    --token $GITHUB_TOKEN \
    --repo $REPO \
    --verbose

# Upload debug-path package
python3 "$UPLOAD_SCRIPT" \
    "$PACKAGES_DIR/debug-path-v1.0.0.tar.gz" \
    --hash-file "$PACKAGES_DIR/debug-path-v1.0.0.tar.gz.sha256" \
    --tag debug-path-v1.0.0 \
    --token $GITHUB_TOKEN \
    --repo $REPO \
    --verbose

Method 3: Bash Script (All at Once)​

#!/bin/bash
# upload_all.sh

cd /Users/rms/Documents/projects/ai-dev-kit
PACKAGES_DIR="packages/frameworks/dist/packages"
UPLOAD_SCRIPT="packages/frameworks/workflow mgt/scripts/upload_to_github_release.py"
REPO="earlution/ai-dev-kit"

# Check token
if [ -z "$GITHUB_TOKEN" ]; then
    echo "❌ Error: GITHUB_TOKEN environment variable not set"
    exit 1
fi

# Array of packages: (package_file, tag)
declare -a PACKAGES=(
    "kanban-v2.1.0.tar.gz:kanban-v2.1.0"
    "workflow-mgt-v2.1.4.tar.gz:workflow-mgt-v2.1.4"
    "numbering-versioning-v2.0.0.tar.gz:numbering-versioning-v2.0.0"
    "doc-lifecycle-v1.0.0.tar.gz:doc-lifecycle-v1.0.0"
    "debug-path-v1.0.0.tar.gz:debug-path-v1.0.0"
)

for package_info in "${PACKAGES[@]}"; do
    IFS=':' read -r package_file tag <<< "$package_info"
    hash_file="${package_file}.sha256"
    
    echo "━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━"
    echo "📤 Uploading: $package_file (tag: $tag)"
    
    if [ ! -f "$PACKAGES_DIR/$package_file" ]; then
        echo "❌ Package not found: $PACKAGES_DIR/$package_file"
        continue
    fi
    
    python3 "$UPLOAD_SCRIPT" \
        "$PACKAGES_DIR/$package_file" \
        --hash-file "$PACKAGES_DIR/$hash_file" \
        --tag "$tag" \
        --token "$GITHUB_TOKEN" \
        --repo "$REPO" \
        --verbose
    
    if [ $? -eq 0 ]; then
        echo "✅ SUCCESS: $package_file"
    else
        echo "❌ FAILED: $package_file"
    fi
    echo ""
done

echo "✅ Upload complete!"

Save as upload_all.sh, make executable, and run:

chmod +x upload_all.sh
./upload_all.sh

Verification​

After uploading, verify packages are on GitHub Releases:

  1. Check Releases Page:

    https://github.com/earlution/ai-dev-kit/releases

  2. Verify Each Release:

    • Each tag should have a release
    • Each release should have 2 assets: .tar.gz and .sha256
    • Download and verify hash matches

  3. Test Installation:

    python3 "packages/frameworks/workflow mgt/scripts/install_package_from_release.py" \
        kanban 2.1.0 \
        --install-dir /tmp/test-install \
        --repo earlution/ai-dev-kit \
        --verbose

Troubleshooting​

Error: Release not found​

  • Solution: Create Git tags first and push them: 
    git tag -a kanban-v2.1.0 -m "Release kanban v2.1.0"
    git push origin kanban-v2.1.0

Error: Authentication failed​

  • Solution: Verify token has repo scope
  • Check token is set: echo $GITHUB_TOKEN
  • Use --token flag to pass directly

Error: Package not found​

  • Solution: Build packages first: 
    python3 "packages/frameworks/workflow mgt/scripts/build_packages_direct.py"

Error: Hash file not found​

  • Solution: Hash file is optional but recommended
  • Package will upload without hash file (less secure)

Next Steps After Upload​

  1. Update Documentation:

    • Add package installation instructions
    • Update framework installation guide
    • Document GitHub Releases as distribution method

  2. Announce Availability:

    • Update README with package installation
    • Create release notes for each framework

  3. Test Installation:

    • Test manual download and installation
    • Test automated installation script
    • Verify hash verification works

Related: