CrowdCent CLI Reference
Authentication
The CLI searches for your API key in this order:
1. CROWDCENT_API_KEY
environment variable
2. .env
file in the current directory with CROWDCENT_API_KEY=your_key_here
Basic Usage
Show help:
Commands
Challenge Management
list-challenges
List all active challenges (JSON output).get-challenge
Get details for a specific challenge.Training Data
list-training-data
List all available training datasets for the challenge.get-latest-training-data
Get metadata for the most recent training dataset.get-training-data
Get metadata for a specific training dataset version.download-training-data
Download a training dataset file.Options:
- -o, --output <path>
- Output location (default: <challenge_slug>_training_v<version>.parquet
)
Example:
Inference Data
list-inference-data
List all inference data periods for the challenge.get-current-inference-data
Get metadata for the currently active inference period.get-inference-data
Get metadata for a specific inference period by date (YYYY-MM-DD).download-inference-data
Download inference features for prediction.Options:
- -o, --output <path>
- Output location (default: <challenge_slug>_inference_<release_date>.parquet
)
Example:
Submissions
list-submissions
List your submissions for a challenge.Options:
- --period <period>
- Filter by period (current
or date in YYYY-MM-DD format)
get-submission
Get details for a specific submission by ID.submit
Submit a prediction file to the current inference period.Requirements:
- File must be Parquet format
- Required columns: id
, pred_1M
, pred_3M
, pred_6M
, pred_9M
, pred_12M
Example:
Error Handling
All errors are printed to stderr. Commands will abort on failure with a non-zero exit code.