@okrapdf/cli

Command-line interface for OkraPDF - extract tables from PDFs.

Installation

npm install -g @okrapdf/cli

Or use with npx:

npx @okrapdf/cli extract invoice.pdf

Quick Start

# Set your API key
export OKRA_API_KEY=okra_xxxxxxxxxxxx

# Extract tables from a PDF
okra extract invoice.pdf -o json

# Ask a question about a document
okra run report.pdf "What is the total revenue?"

Get your API key from okrapdf.com/settings/api-keys.

Configuration

The CLI looks for your API key in this order:

1. OKRA_API_KEY environment variable
2. .env file in current directory
3. .okra file in current directory
4. ~/.okra file in home directory
5. ~/.config/okrapdf/config.json

Example .env or .okra file:

OKRA_API_KEY=okra_xxxxxxxxxxxx
OKRA_BASE_URL=https://okrapdf.com

Commands

Shortcuts (Most Common)

okra extract <file>          # Upload + extract + wait (all-in-one)
okra run <file> "question"   # Extract + ask question

Documents

okra docs list               # List all documents
okra docs upload <file>      # Upload a PDF
okra docs get <uuid>         # Get document details
okra docs delete <uuid>      # Delete a document

Jobs

okra jobs list               # List extraction jobs
okra jobs create <file>      # Create extraction job
okra jobs get <job-id>       # Get job status
okra jobs wait <job-id>      # Wait for completion
okra jobs results <job-id>   # Get extraction results

Tables

okra tables list <doc-uuid>  # List extracted tables
okra tables get <table-id>   # Get table content
okra tables export <id>      # Export to CSV/JSON

Interactive Chat

okra chat <document-uuid>    # Interactive document Q&A

Output Formats

All commands support -o, --output:

• table (default) - Human-readable tables
• json - Machine-readable JSON
• csv - CSV format

# JSON for scripting
okra jobs list -o json | jq '.[].id'

# Quiet mode for piping
okra extract doc.pdf -o json -q > results.json

For AI Agents

The CLI is designed to be agent-friendly:

# Get tables as JSON (for building presentations, reports)
okra extract document.pdf --json --quiet

# Get document list as JSON for processing
okra docs list -o json | jq '.[].uuid'

# Extract with specific processor
okra jobs create document.pdf -p gemini --wait -o json

Environment Variables

| Variable | Description | |----------|-------------| | OKRA_API_KEY | API key (required) | | OKRA_BASE_URL | Custom API URL (for self-hosted) | | OKRA_OUTPUT_FORMAT | Default output format |

Examples

Batch Processing

for pdf in *.pdf; do
  okra extract "$pdf" -o json > "${pdf%.pdf}.json"
done

CI/CD Integration

# Extract and check for tables
RESULT=$(okra extract report.pdf -o json -q)
TABLE_COUNT=$(echo "$RESULT" | jq '.tables | length')
echo "Found $TABLE_COUNT tables"

Exit Codes

| Code | Meaning | |------|---------| | 0 | Success | | 1 | General error | | 2 | Invalid arguments | | 3 | Authentication error | | 4 | Resource not found | | 5 | Rate limited |

Development

git clone https://github.com/okrapdf/cli
cd cli
npm install
npm run build
npm link
okra --help

License

MIT - see LICENSE

Links

• OkraPDF
• Documentation
• API Reference