RESEARCH

Research that would take weeks — done overnight

Multiple agents collaborate across sessions to conduct deep research — reviewing thousands of sources, synthesizing findings, and delivering comprehensive reports. From economic impact studies to market analysis, this is research at a scale humans can't match.

What you get

Multi-agent research at scale

A lead agent delegates to specialized sub-agents — each covering a different research domain simultaneously. What would take a team of analysts weeks happens in hours.

Thousands of sources reviewed

Academic papers, industry reports, news articles, and social media — agents cast a wide net across 1,200+ sources to ensure comprehensive coverage.

Cross-referenced synthesis

Findings are validated across multiple sources. Conflicting data is flagged, consensus views are highlighted, and outlier perspectives are noted.

Structured deliverable

Executive summary for stakeholders plus a full report with citations, data tables, and methodology notes — ready to share or publish.

Iterative depth

Follow-up sessions can drill deeper into specific areas. The agent builds on previous research, refining findings and exploring new angles without starting from scratch.

Done by an agent. Zero human effort.

Kick it off before you leave for the day. The lead agent spawns sub-agents, coordinates research across domains, and delivers a full report to Slack by morning.

Research Agent · Deep Research Complete
> Spawned 5 sub-agents for parallel research
> Agent 1 covering market size & growth trends
> Agent 2 covering competitive landscape & positioning
> Agent 3 covering regulatory & policy environment
> Agent 4 covering technology trends & innovation
> Agent 5 covering consumer sentiment & social signals

Research complete:
> Sources reviewed 1,247 papers, articles, and reports
> Key findings synthesized across 5 domains
> Cross-referenced 83 claims across multiple sources
> Final report compiled — 47 pages with citations
> Executive summary generated — 2 pages

✓ Full report saved as artifact
✓ Executive summary posted to #research on Slack
Duration: 4h 12m · 5 sub-agents used
HOW TO SET IT UP

For CRHQ clients

1

Enable the required skills

Make sure these skills are active on your satellite:

2

Define your research scope

Example prompt:

"Conduct deep research on the economic impact of generative AI on the financial services industry. Use delegation to spawn sub-agents covering market size, regulatory landscape, competitive dynamics, technology adoption curves, and consumer sentiment. Review academic papers, industry reports, and news. Cross-reference findings and compile a full report with executive summary. Post the summary to #research on Slack."
3

Let it run (or schedule it)

Deep research can take 2–8 hours depending on scope. Start it before you leave for the day, or use the schedule skill to kick it off overnight. Results arrive in Slack when complete.

ADVANCED TIPS

Get more out of this workflow

Use project documents as a research workspace

Tell the agent to use project documents as a shared workspace where sub-agents deposit their findings. This creates a structured research trail that the lead agent can synthesize — and that you can reference in follow-up sessions for iterative depth.

Break complex topics into sub-questions

The more specific your research brief, the better the delegation. Instead of "research AI in healthcare," break it into targeted sub-questions: market size, regulatory hurdles, adoption rates, key players, and patient outcomes. Each sub-question becomes a focused agent task.

Combine with Google Trends for real-time signals

Academic papers and industry reports show what happened. Google Trends shows what's happening now. Pair them together and your research captures both historical analysis and real-time market momentum.

google-trends

Use Grok for social media sentiment

Add the grok skill to pull X/social media sentiment into your research. This is especially powerful for consumer-facing topics where public opinion, trending conversations, and influencer takes shape the narrative.

grok
Go deep.
$299/month per satellite. Bring your own AI model. Deploy in 90 seconds.