company-investigation
Pre-meeting company research — paste a URL, get a scannable HTML report.
A Claude Cowork skill that dispatches parallel AI research agents across 10 dimensions and delivers a self-contained HTML report with source credibility badges — ready before your next meeting.
What you get
/investigate https://notion.so
┌─────────────────────────────────────────────────────┐
│ Notion │
│ "The all-in-one workspace" 2026-03-26 │
├─────────────────────────────────────────────────────┤
│ Founded │ HQ │ Employees │ Funding │
│ 2013 │ San Francisco│ ~800 │ $343M │
├─────────────────────────────────────────────────────┤
│ EXECUTIVE SUMMARY │
│ • Valued at $10B (2021), PLG → enterprise push │
│ • Notion AI at $10/member/mo — new revenue stream │
│ • Main threat: Microsoft Loop bundled in M365 │
├─────────────────────────────────────────────────────┤
│ ▸ 1. Company Overview [Verified] │
│ ▸ 2. Product / Service [Verified] │
│ ▸ 3. Team / Leadership [Reported] │
│ ▸ 4. Business Model [Verified] │
│ ▸ 5. Market & Competitors [Reported] │
│ ▸ 6. Financials & Funding [Verified] │
│ ▸ 7. Tech Stack / Engineering [Unconfirmed] │
│ ▸ 8. News & Sentiment [Reported] │
│ ▸ 9. Red Flags / Risks [Reported] │
│ ▸ 10. AI Moat Analysis [Reported] │
├─────────────────────────────────────────────────────┤
│ AI MOAT SCORE: Low [ Medium ] High Strong │
├─────────────────────────────────────────────────────┤
│ TALKING POINTS │
│ 1. "How is Notion AI attach rate tracking?" │
│ 2. "How are you handling the Loop bundling risk?" │
│ 3. "What % of revenue is now enterprise?" │
├─────────────────────────────────────────────────────┤
│ SOURCES │
│ [Verified] notion.so, Crunchbase + Bloomberg │
│ [Reported] TechCrunch, G2, Glassdoor │
│ [Unconfirmed] LinkedIn estimates, Reddit signals │
└─────────────────────────────────────────────────────┘
How it works
User: /investigate https://company.com
│
▼
┌─────────────────────┐
│ STEP 1: QUICK │ WebFetch company URL
│ SCAN │ Extract: name, tagline, subpages
│ │ Show 10 sections → ask to adjust
└────────┬────────────┘
│ user confirms "go"
▼
┌────────────────────────────────────────────────┐
│ STEP 2: DEEP RESEARCH │
│ (4 agents run in parallel) │
│ │
│ ┌───────────────┐ ┌───────────────────────┐ │
│ │ Company Agent │ │ Market Agent │ │
│ │ Sections 1–4 │ │ Section 5 │ │
│ │ (company URL │ │ (G2, Capterra, │ │
│ │ + subpages) │ │ web search) │ │
│ └───────┬───────┘ └──────────┬────────────┘ │
│ │ │ │
│ ┌───────┴───────┐ ┌──────────┴────────────┐ │
│ │ Financial / │ │ AI Moat Agent │ │
│ │ News Agent │ │ Sections 9–10 │ │
│ │ Sections 6–8 │ │ (job postings, │ │
│ │ (Crunchbase, │ │ GitHub, patents) │ │
│ │ news, G2) │ │ │ │
│ └───────┬───────┘ └──────────┬────────────┘ │
│ └──────────┬───────────┘ │
└───────────────────────────────────────────────-─┘
│
▼
┌────────────────────────┐
│ STEP 3: SYNTHESIS │
│ │
│ 1. Merge all findings │
│ 2. Resolve conflicts │
│ 3. Tag credibility │
│ 4. Generate HTML │
│ 5. Save + open report │
└────────────────────────┘
│
▼
notion-investigation-2026-03-26.html
Source credibility
Every claim in the report is tagged:
┌────────────────┬─────────────────────────────────────────────┐
│ Badge │ Criteria │
├────────────────┼─────────────────────────────────────────────┤
│ [Verified] │ Company's own URL, or 2+ independent sources │
│ [Reported] │ Single reputable source (Reuters, G2, etc.) │
│ [Unconfirmed] │ Social media, forums, inferred/estimated │
└────────────────┴─────────────────────────────────────────────┘
Conflicts between sources are flagged explicitly — never silently resolved.
Install
Claude Cowork (desktop)
anysiteio
ameya85
firecrawl
fivetaku
hoangsonww
genli-ai