parallel-cli

Run structured web research and extraction jobs via terminal CLI with JSON output.

Updated May 5, 2026
One-click install
npx skills add https://github.com/Z43L/zeus-agent --skill parallel-cli-z43l
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parallel-cli
Source: https://github.com/Z43L/zeus-agent/tree/main/optional-skills/research/parallel-cli
Command: npx skills add https://github.com/Z43L/zeus-agent --skill parallel-cli-z43l

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Parallel-cli solves the problem of turning messy, multi-step web research into structured, agent-ready outputs—so you can search, extract, research, enrich, discover entities, and monitor sources without constantly switching tools or manually post-processing results.

Core Features & Use Cases

  • Agent-native web research workflows: launch deep research jobs, poll for progress, and retrieve final reports with consistent structure.
  • Extraction and deep content retrieval: fetch and extract clean text or targeted information from specific URLs for accurate downstream summaries.
  • Enrichment, FindAll, and monitoring: enrich datasets with additional fields, generate discovered entity lists (datasets), and set up change monitoring for ongoing tracking.

Quick Start

Ask: "Use parallel-cli to run research on 'Compare the leading AI coding agents by pricing, model support, and enterprise controls' and return the results as JSON."

Frequently Asked Questions about parallel-cli

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

FAQPage Schema
How do I get structured JSON output from web research for agent workflows?

To get structured JSON output from web research, you can run terminal-first CLI commands with the --json flag to search, extract, and enrich data into agent-ready formats without manual post-processing.

Can I run async web research jobs without waiting for them to finish?

Yes, you can launch async web research jobs using the --no-wait flag to execute non-interactively, and then use status polling to check progress and retrieve final reports when ready.

What's the best way to enrich datasets with discovered entities from web sources?

Dataset enrichment with discovered entities is handled by generating targeted entity lists and adding additional fields from web sources, turning messy, multi-step research into structured, agent-ready outputs.

Does this web research tool support follow-up context chaining for multi-step investigations?

Yes, follow-up context chaining supports multi-step investigations by linking sequential search, extraction, and deep research jobs, allowing you to build datasets and cite-from-returned-URLs accurately.

How do I monitor web sources for ongoing changes using a CLI?

You can set up source change monitoring via the terminal-first CLI to track ongoing updates, using async jobs with polling to maintain continuous tracking without interactive intervention.

When should I use async polling instead of waiting for deep research to complete?

Use async polling with --no-wait for long-running deep research operations, allowing the terminal to remain free while the job processes and returning structured JSON results upon status completion.