parallel-cli

Runs Parallel CLI web search, extraction, research, enrichment, FindAll and monitoring workflows with JSON output and async job management via --no-wait, status, poll, and --previous-interaction-id chaining.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kotakbiasa/hermes-agent --skill parallel-cli-kotakbiasa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parallel-cli
Source: https://github.com/kotakbiasa/hermes-agent/tree/main/optional-skills/research/parallel-cli
Command: npx skills add https://github.com/kotakbiasa/hermes-agent --skill parallel-cli-kotakbiasa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This vendor skill provides a formal pathway to run Parallel CLI for advanced web search, extraction, deep research, data enrichment, FindAll-style discovery, and monitoring, enabling non-Hermes workflows when needed.

Core Features & Use Cases

  • Vendor-specific workflows for web search, extraction, research, enrichment, FindAll, and monitoring via Parallel CLI.
  • JSON-output by default and non-interactive execution, with async job management using --no-wait, status, and poll.
  • Context chaining support through --previous-interaction-id to chain related tasks and maintain session continuity.
  • Use cases include long-running research pipelines, entity discovery, and ongoing monitoring needs that extend Hermes native capabilities.

Quick Start

Run a vendor-enabled search with parallel-cli search "AI coding agents" --json to see structured results.

Frequently Asked Questions about parallel-cli

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

FAQPage Schema
How do I run async web search and data enrichment pipelines non-interactively?

Async web search and data enrichment pipelines run non-interactively using Parallel CLI, which executes jobs with JSON output by default and manages them using flags like --no-wait, status, and poll. This enables automated, long-running research workflows without active terminal sessions.

What is context chaining for multi-step research tasks?

Context chaining for multi-step research tasks maintains session continuity across separate executions by passing the --previous-interaction-id flag. This links related queries together, allowing subsequent async jobs to inherit and build upon the data context of prior steps in the pipeline.

How do I monitor websites and extract structured JSON data continuously?

Website monitoring and structured JSON data extraction are handled by Parallel CLI's built-in monitor and extraction features. The CLI continuously tracks target sources and automatically returns extracted findings as JSON output, suitable for ongoing entity discovery and data enrichment workflows.

Does Parallel CLI support long-running deep research workflows with job polling?

Parallel CLI supports long-running deep research workflows through native async job polling. Users can initiate a research task with --no-wait and subsequently use the status and poll commands to check execution progress and retrieve final JSON results without blocking the terminal.

How do I chain multiple web search queries to maintain session continuity?

You chain multiple web search queries to maintain session continuity by utilizing the --previous-interaction-id parameter in Parallel CLI. This passes the context from a prior interaction to the new execution, ensuring the subsequent search operation builds upon previous findings.

Can I use Parallel CLI for entity discovery without interactive prompts?

Parallel CLI is designed for non-interactive execution, making it suitable for entity discovery and FindAll-style discovery workflows. It operates natively in automated pipelines by defaulting to JSON output and eliminating interactive prompts during web search and data extraction processes.