parallel-web/parallel-agent-skills - Setup Guide¶
Source: parallel-web/parallel-agent-skills (9.2K installs for parallel-monitor) Category: Web Intelligence / Agent Tools Harnesses: Claude Code, Codex CLI, Cursor, GitHub Copilot, Cline
Web search, content extraction, deep research, and data enrichment for AI coding agents. Built on Parallel infrastructure. Provides agents with real-time web access for research, monitoring, and data gathering - without the overhead of managing browser instances.
Installation¶
Claude Code (Recommended for Hermes)¶
# Add the marketplace
/plugin marketplace add parallel-web/parallel-agent-skills
# Install the Parallel plugin
/plugin install parallel
# RESTART Claude Code before continuing!
# Setup CLI + auth
/parallel:parallel-cli-setup
Skills.sh (Universal)¶
# Install all Parallel skills into any compatible agent
npx skills add parallel-web/parallel-agent-skills@parallel-monitor
OpenAI Codex¶
$skill-installer parallel-web/parallel-agent-skills
/parallel:parallel-cli-setup
Verify:
parallel-cli status
# Should show: Authenticated ✓, Balance: $X.XX
Prerequisites¶
| Requirement | Details |
|---|---|
| parallel-cli | Install via skill setup or directly: npm i -g parallel-cli |
| API Key | Get from parallel.ai - free tier available |
| Hermes Agent | Any version with Claude Code plugin support |
| Funded Account | Required for web extraction (pay-as-you-go) |
Install CLI if missing:
npm install -g parallel-cli
parallel-cli auth login
Key Skills¶
| Skill | Installs | Purpose |
|---|---|---|
| parallel-monitor | 9.2K | Monitor websites for changes, trigger on conditions |
| parallel-search | Included | Web search with structured results |
| parallel-extract | Included | Content extraction from URLs |
| parallel-deep-research | Included | Multi-source research synthesis |
| parallel-data-enrich | Included | Enrich data with web sources |
| parallel-cli-setup | Included | One-time CLI install + auth setup |
Common Workflows for CorpusIQ¶
Competitive Monitoring¶
# Monitor competitor pricing pages for changes
/parallel:monitor \
--url "https://competitor.com/pricing" \
--check-interval 6h \
--alert-on "price change" \
--notify telegram
# Monitor competitor blog for new posts
/parallel:monitor \
--url "https://competitor.com/blog" \
--check-interval 24h \
--alert-on "new article" \
--extract-title-and-summary
Market Research¶
# Deep research on a market segment
/parallel:deep-research \
--topic "AI agent orchestration platforms 2026" \
--sources 10 \
--output market-research.md
# Search for specific intel
/parallel:search \
--query "CorpusIQ competitors funding announcements 2026" \
--max-results 15 \
--output competitor-intel.json
Content Research¶
# Extract content from multiple URLs for blog research
/parallel:extract \
--urls "https://example1.com,https://example2.com,https://example3.com" \
--format markdown \
--output research-notes.md
# Enrich lead data with web sources
/parallel:data-enrich \
--input leads.csv \
--columns company_name,website \
--enrich company_size,industry,funding \
--output enriched-leads.csv
Social Media Monitoring¶
# Monitor brand mentions (combined with web search)
/parallel:monitor \
--search "CorpusIQ" \
--sources web,news,social \
--check-interval 12h \
--output-dir ./brand-monitoring/
Comparison: parallel-web vs Other Web Tools¶
| Feature | parallel-web | Firecrawl | agent-browser | Apify |
|---|---|---|---|---|
| Web search | ✅ Structured | ✅ Structured | ❌ (browser only) | ✅ via Actors |
| Content extraction | ✅ Parallel infra | ✅ Single-page | ✅ Accessibility tree | ✅ via Actors |
| Monitoring | ✅ Built-in | ✅ Built-in | ❌ Manual | ✅ via Schedules |
| Deep research | ✅ Multi-source | ✅ Workflows | ❌ | ✅ via Actors |
| Data enrichment | ✅ Built-in | ❌ | ❌ | ✅ via Actors |
| Install base | 9.2K | 120K | 553K | 2.5K |
| Hermes integration | Via Claude Code plugin | Via CLI/MCP | Via MCP | Via API |
| Pricing | Pay-as-you-go | Subscription | Free (OSS) | Pay-per-use |
Recommendation: Use parallel-web for monitoring and deep research. Use agent-browser for one-off interactive browsing. Use Firecrawl for bulk extraction. They complement each other.
Integration with Hermes Agent¶
Direct CLI Usage¶
# From Hermes terminal
parallel-cli search "hermes agent plugins" --max-results 10
parallel-cli extract "https://example.com" --format markdown
parallel-cli monitor --url "https://target.com" --interval 6h
Via Claude Code Plugin¶
# Interactive within Hermes Claude Code session
/parallel:search --query "latest AI agent frameworks" --max 5
/parallel:deep-research --topic "agent memory systems comparison"
/parallel:monitor --url "https://competitor.com" --interval 12h
Troubleshooting¶
| Symptom | Cause | Fix |
|---|---|---|
/parallel: command not found |
Plugin not installed | /plugin install parallel then restart |
parallel-cli: not found |
CLI not installed | npm install -g parallel-cli |
Authentication failed |
API key expired/missing | parallel-cli auth login |
Insufficient balance |
Account needs funds | Add funds at parallel.ai |
| Monitor not triggering | Interval too long / URL changed | Check monitor status: parallel-cli monitor list |
| Timeout on extract | Page too large / JS-heavy | Increase timeout: --timeout 30 |
Integration with CorpusIQ¶
CorpusIQ is the connector layer that lets AI assistants read the systems a business already runs on - accounting, payments, CRM, email, files and analytics - and answer with the source named. It is read-only by design: it retrieves data and never writes to, or acts in, a connected system, which is what makes it safe to point at live accounts.
This page is part of the CorpusIQ directory, the reference we maintain for the MCP server and agent-skill ecosystem. Run alongside CorpusIQ, a third-party server like this one can be combined with a company's own business data in a single answer.
See Also¶
- Firecrawl Workflows Setup - Web scraping and SEO audits (120K installs)
- agent-browser Setup - Accessibility-tree browser automation (553K installs)
- Apify Agent Skills Setup - Web scraping Actors
- wshobson Agents Marketplace - 94-plugin agent marketplace