parallel-cli

Runs vendor-specific web search and extraction workflows with JSON outputs.

2|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/CUexter/hermes-agent --skill parallel-cli-cuexter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: parallel-cli
Source: https://github.com/CUexter/hermes-agent/tree/main/skills/research/parallel-cli
Command: npx skills add https://github.com/CUexter/hermes-agent --skill parallel-cli-cuexter

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Parallel CLI provides a vendor-specific workflow for web search, extraction, deep research, enrichment, and monitoring with JSON-first outputs that integrate into agent pipelines, enabling robust, structured data collection.

Core Features & Use Cases

  • Vendor-specific web search, extraction, deep research, enrichment, and monitoring workflows.
  • JSON-first outputs for machine-readable results and easy integration into automations.
  • Async long-running tasks with status/poll, context chaining, and multi-step pipelines.

Quick Start

Ask the agent to run a vendor-specific web search and return structured results in JSON for immediate processing.

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 automated web search for agent workflows?

Automated web search for agent workflows can return structured JSON output by enforcing a JSON-first response format, ensuring machine-readable results for immediate processing in non-interactive pipelines. This enables robust data collection without manual parsing.

Can I run async long-running research tasks and check their status later?

Yes, async long-running research tasks are supported through a status and polling mechanism. You can initiate a search or extraction task non-interactively and retrieve its results later, enabling deep research across multiple sources without blocking execution.

How do I enrich extracted web data and chain context across multiple steps?

You can enrich extracted web data and chain context using a previous-interaction-id. This allows modular workflows to pass context between search, extract, and enrich steps, building multi-step pipelines for comprehensive research and monitoring.

Does this structured web search tool work without user interaction by default?

Yes, this structured web search tool operates in a non-interactive mode by default. It is designed for agent-driven environments, allowing automated extraction and monitoring to run without requiring manual user input during execution.

What is the best way to monitor multiple web sources for changes automatically?

Monitoring multiple web sources for changes is handled through dedicated monitor workflows that poll asynchronously and return JSON. This approach provides structured updates suitable for automated tracking and agent-driven data collection pipelines.