parallel-cli

Runs web search, extraction, deep research, enrichment, and monitoring via the Parallel CLI.

1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/kaminocorp/hermes-alpha-hunter --skill parallel-cli-kaminocorp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: parallel-cli
Source: https://github.com/kaminocorp/hermes-alpha-hunter/tree/main/skills/research/parallel-cli
Command: npx skills add https://github.com/kaminocorp/hermes-alpha-hunter --skill parallel-cli-kaminocorp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? This Skill gives agents a terminal-native way to use Parallel's vendor stack for web search, URL extraction, deep research, data enrichment, entity discovery, and change monitoring, with JSON output and non-interactive async workflows. ## Core Features & Use Cases - Search and Extraction: Run structured web searches with domain and date filters, and extract clean markdown content from URLs with optional objectives. - Deep Research and Enrichment: Launch async research jobs with processor tiers, poll for results, and enrich CSV or JSON datasets with web-derived columns. - FindAll and Monitoring: Discover entity datasets at web scale and set up recurring monitors that detect changes over time. - Use Case: A user asks to compare AI coding agents by pricing and enterprise controls. The agent launches a research run with --no-wait --json, continues other work, polls the run ID later, and summarizes the report with citations from returned sources. ## Quick Start Ask the agent to use parallel-cli to search for a topic with JSON output, for example by requesting a search on the latest funding news for a company with cited sources.

Frequently Asked Questions about parallel-cli

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

FAQPage Schema
How do I run a web search with parallel-cli?▼

Run parallel-cli search followed by your query and the --json flag for structured output. You can narrow results with --include-domains, --exclude-domains, --after-date, and --max-results flags.

How do I run async deep research with parallel-cli?▼

Use parallel-cli research run with your question, a processor tier like core or ultra, and --no-wait --json to launch the job. Then poll the returned run ID with research status or research poll until the report completes.

When should I use parallel-cli instead of built-in web search?▼

Use parallel-cli when the user explicitly requests Parallel or needs vendor-specific features like enrichment, FindAll entity discovery, or monitoring. For quick one-off lookups, native web_search and web_extract tools are preferred.

How do I authenticate parallel-cli in headless or CI environments?▼

Set the PARALLEL_API_KEY environment variable, or run parallel-cli login --device for headless device-based login. Verify the current auth state at any time with parallel-cli auth.

What do parallel-cli exit codes mean?▼

Exit code 0 means success, 2 indicates bad input, 3 is an auth error, 4 is an API error, and 5 is a timeout. For auth errors, check parallel-cli auth and confirm your API key or login status.