parallel-cli

Run parallel web research workflows with JSON output and async polling.

Updated May 4, 2026

Search Tags:#monitoring #web search #data enrichment #json output #deep research #entity discovery #parallel-cli

Ready to run

npx skills add https://github.com/JamesFincher/gengar --skill parallel-cli-jamesfincher

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: parallel-cli
Source: https://github.com/JamesFincher/gengar/tree/main/optional-skills/research/parallel-cli
Command: npx skills add https://github.com/JamesFincher/gengar --skill parallel-cli-jamesfincher

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of producing richer, structured web research results—search, extraction, deep research, enrichment, entity discovery, and monitoring—without manual copy/paste and with agent-friendly non-interactive execution.

Core Features & Use Cases

Agent-native parallel research workflows: Launch and poll long-running research jobs, then continue follow-ups using prior interaction IDs.
Structured outputs for automation: Prefer JSON output via --json to reliably chain results into later steps or external tools.
Deep enrichment and discovery: Use enrichment to add inferred columns to tabular data and FindAll to produce a discovered dataset rather than a short answer.

Quick Start

Use parallel-cli to research a topic with machine-readable output for downstream summarization by running the command: parallel-cli search "What are the latest enterprise AI coding agent controls?" --json.

Frequently Asked Questions about parallel-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run deep web research and get JSON output for automation pipelines?▼

You can run deep web research with structured JSON output by using the `--json` flag with your search query. This ensures machine-readable results that can be reliably chained into downstream automation steps or external tools without manual formatting.

Can I launch long-running research jobs asynchronously and check their status later?▼

Yes, you can launch long-running research jobs asynchronously using the `--no-wait` flag. You can then poll the job status later to retrieve results, enabling non-interactive terminal workflows for extensive data extraction and deep research tasks.

How do I enrich tabular data and discover new entities during web research?▼

You can enrich tabular data by using the enrichment feature to add inferred columns, and use the FindAll function for entity discovery. This produces a comprehensive discovered dataset rather than just a short answer to your research query.

What is the best way to chain follow-up research queries using prior search context?▼

You can chain follow-up research queries by referencing prior interaction IDs from previous parallel-cli executions. This context chaining allows you to continue deeper research workflows without losing the discovered data or entity relationships from earlier steps.

Does parallel-cli require any external dependencies or components to run web searches?▼

No, parallel-cli does not require any external dependencies or components to run. It is a self-contained terminal-native tool designed for agent-native web search, extraction, and monitoring workflows without additional environment setup.

Why should I use a CLI tool for web research instead of manual browser searches?▼

Using a CLI tool eliminates manual copy-paste by producing richer, structured web research results directly in your terminal. It supports agent-friendly non-interactive execution and ensures citations only include returned URLs, preventing invented sources.