PipelineIQ

AI-powered CI pipeline analyzer - identify inefficiencies and optimize your pipelines.

Features

• 🔍 8 Built-in Analysis Rules - Detect caching issues, parallelization opportunities, security concerns, and more
• 🤖 AI-Powered Suggestions - Get intelligent recommendations using Claude AI
• 📊 Multiple Output Formats - Terminal, JSON, and Markdown reports
• 🎯 Harness CI Support - Built for Harness CI pipelines (more platforms coming)
• ⚡ Fast Analysis - Analyze pipelines in milliseconds

Installation

# Using Poetry (recommended)
poetry install

# Or using pip
pip install -e .

Quick Start

# Analyze a Harness CI pipeline
pipelineiq analyze pipeline.yaml --platform harness

# Get JSON output for CI integration
pipelineiq analyze pipeline.yaml --platform harness --format json

# Enable AI suggestions (requires ANTHROPIC_API_KEY)
export ANTHROPIC_API_KEY="your-api-key"
pipelineiq analyze pipeline.yaml --platform harness --ai

# List all available rules
pipelineiq list-rules

# Explain a specific rule
pipelineiq explain cache-dependencies

CLI Commands

analyze

Analyze a CI pipeline file for optimization opportunities.

pipelineiq analyze <path> --platform <platform> [options]

Required:

• <path> - Path to the pipeline file
• --platform, -p - CI platform (currently: harness)

Options:

• --format, -f - Output format: terminal (default), json, markdown
• --severity, -s - Minimum severity: critical, high, medium, low
• --rules, -r - Comma-separated rule IDs to run
• --ai - Enable AI-powered suggestions
• --output, -o - Write output to file
• --verbose, -v - Verbose output

Exit Codes:

• 0 - No findings
• 1 - Findings found
• 2 - Critical issues (score < 50)
• 3 - Parse error
• 4 - File not found
• 5 - Other error

list-rules

List all available analysis rules.

pipelineiq list-rules

explain

Explain a specific analysis rule.

pipelineiq explain <rule-id>

version

Show version information.

pipelineiq version

Analysis Rules

Rule ID	Severity	Description
cache-dependencies	HIGH	Detects package manager installs without caching
cache-docker-layers	MEDIUM	Docker builds without layer caching
parallel-stages	HIGH	Independent stages that could run in parallel
parallel-steps	MEDIUM	Independent steps within a job
missing-timeout	MEDIUM	Jobs/stages without timeout configuration
redundant-clone	HIGH	Multiple git clone operations
pinned-versions	HIGH	Unpinned image/plugin versions
resource-sizing	LOW	Mis-sized compute resources

Example Output

╭─────────────────── Analysis ───────────────────╮
│ PipelineIQ Analysis Report                     │
│ Pipeline: My CI Pipeline                       │
│ Platform: harness                              │
│ File: pipeline.yaml                            │
╰────────────────────────────────────────────────╯
╭─────────────────── Summary ────────────────────╮
│        Score:  78/100                          │
│     Findings:  4                               │
│    Breakdown:  🟠 1 high | 🟡 2 medium | 🔵 1 low │
│ Est. Savings:  2-8 minutes per run             │
╰────────────────────────────────────────────────╯

AI Integration

PipelineIQ integrates with Claude AI to provide intelligent suggestions:

# Set your API key
export ANTHROPIC_API_KEY="your-api-key"

# Run with AI suggestions
pipelineiq analyze pipeline.yaml --platform harness --ai

AI features:

• Overall Suggestions - Prioritized recommendations based on findings
• Finding Explanations - Detailed explanations of why issues matter
• Fix Generation - Suggested YAML fixes (coming soon)

Development

# Install dependencies
poetry install

# Run tests
poetry run pytest

# Run with coverage
poetry run pytest --cov=pipelineiq

# Type checking
poetry run mypy src/

# Linting
poetry run ruff check src/

Project Structure

pipelineiq/
├── src/pipelineiq/
│   ├── cli/          # Typer CLI application
│   ├── core/         # Analysis engine and DAG builder
│   ├── models/       # Pydantic data models
│   ├── parsers/      # Platform-specific parsers
│   ├── analyzers/    # Analysis rules
│   ├── reporters/    # Output formatters
│   └── ai/           # Claude AI integration
├── tests/            # Test suite
└── pyproject.toml    # Project configuration

Roadmap

• GitHub Actions support
• GitLab CI support
• CircleCI support
• Web UI
• VS Code extension
• More analysis rules

License

MIT License - see LICENSE for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.