Agent Role
Research Analyst
Briefings, not busywork.
What they own
Market research, competitive intelligence, news monitoring, and report compilation. Your agent runs the information gathering so you get the insight without the grind — a morning brief waiting when you wake up, on-demand research when you need it, and a weekly digest that filters the noise before it reaches you.
The best decisions are made with current information. But gathering that information — scanning sources, synthesizing reports, tracking what changed — is work that doesn't require your judgment. Your agent does the gathering. You do the deciding.
What changes for you
6:00 AM
Overnight research run completes.
Five source types scanned across eight cycles. Three competitor moves, one pricing change, one funding announcement. Two-paragraph brief ready in Slack.
You knew what mattered before your first meeting.
10:30 AM
"What's the competitive landscape for our pricing tier?"
12 sources synthesized. Competitors mapped by tier and positioning. One market gap identified. Two-paragraph brief with a recommended next step.
An analyst would have taken two days.
Friday afternoon
Weekly digest delivers automatically.
Everything in your industry this week, filtered to what affects your decisions. Three items need your attention. The rest ignored.
You spent the week building. Your agent spent it watching.
What they can do
How to bring them on board
Provision your vessel
Your research data and competitive intelligence on isolated infrastructure.
Set your LLM provider key
Add your Anthropic, OpenAI, or Google API key. This is what powers the agent — without it, nothing runs. Takes 30 seconds in the Studio.
Connect your channels
Slack, email, or wherever you want briefings delivered.
Define your research scope
Which competitors, which industries, which topics. The agent focuses on what matters to you.
Schedule your first briefing
Set up a daily or weekly competitive scan. Wake up to intelligence, not an empty dashboard.
Skills to install
These ClawHub skills unlock the capabilities above. Scan each before installing — the trust scanner catches known malicious skills.
Web Search
On-demand research, competitive scans, news monitoring
clawhub install web-searchScan before installing →RSS Reader
Subscribe to industry feeds, track competitor blogs
clawhub install rss-readerScan before installing →PDF Reader
Summarize analyst reports, whitepapers, earnings filings
clawhub install pdf-readerScan before installing →Reddit Search
Community signal — what practitioners are actually doing
clawhub install reddit-searchScan before installing →Hacker News
Tech and startup signal — funding, launches, discussion
clawhub install hacker-newsScan before installing →Common questions
What sources does the research agent monitor?
Web search (news, competitor sites), RSS feeds (industry publications), Reddit communities, Hacker News, and PDF documents (analyst reports, earnings filings, whitepapers) — depending on which skills you install.
How often does it run and deliver briefings?
You set the schedule. Most users configure a daily 6am brief and a Friday afternoon digest. On-demand research runs whenever you ask — typically returning results in under two minutes.
Can it track specific competitors or topics?
Yes. During setup you define the scope: which companies, which industries, which keywords. The agent focuses on what you tell it matters and filters the rest.
What does a typical briefing look like?
A short Slack message or email with 3–5 bullet points — what changed, why it matters, what to watch. Long-form reports available on request. Signal without noise.
Your agent surfaces. You interpret.
Briefings and competitive summaries your agent compiles are starting points for your analysis, not finished conclusions. Your agent processes volume. You apply context, judgment, and domain knowledge to turn information into decisions.
