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 |
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