parallel-cli

You automate web research by running parallel search, extraction, and enrichment workflows that return machine-readable JSON. This replaces slow manual browsing with structured, non-interactive outputs for downstream use.

CSV and JSON-based enrichment is handled by launching parallel data enrichment workflows. The tool processes your input sources asynchronously and appends extracted information directly into the structured JSON output.

Async web monitoring jobs are controlled explicitly using the --no-wait flag. You can launch long-running change tracking tasks and poll their status later to retrieve machine-readable JSON results without blocking execution.

Large-scale entity discovery is best handled by launching parallel search and findall workflows. The tool automates information extraction across user-defined sources and returns all discovered entities in a structured JSON format.

Citation discipline is maintained by strictly using only the URLs returned in the CLI output. When summarizing deep research results, you must reference these explicit URLs to ensure all extracted information is verifiable.

Async deep research workflows require non-interactive execution and explicit job polling to manage long-running tasks. You cannot retrieve intermediate results without checking the job status, and outputs are strictly formatted as JSON.