Available Context & Tools

@_platform-references/org-variables.md @_platform-references/capabilities.md

Company Research Report

Goal

Produce a comprehensive, human-readable intelligence report on a target company. The report should feel like a professional analyst briefing — factual, sourced, and actionable for sales teams.

Required Capabilities

Web Search: To research company information across multiple sources (routed to search-capable model)

Inputs

company_name: Name of the company to research (required)
company_website: Company website URL or domain (if known, speeds up research significantly)
founder_name: Founder or CEO name (helps disambiguate when multiple companies share a name)
industry: Industry context (helps focus research on the right company and relevant metrics)

Research Methodology

Phase 1: Discovery (run searches in parallel)

Execute these searches simultaneously:

Primary: "[Company Name]" company — official site, Wikipedia, Crunchbase
Product/service: "[Company Name]" product OR platform OR service — what they sell
Leadership: "[Company Name]" founder OR CEO OR "co-founder" — who runs it
News: "[Company Name]" news OR announcement OR funding OR acquisition — recent activity
Reviews: "[Company Name]" reviews OR Trustpilot OR G2 OR Capterra — market reputation

If a domain is known, also: 6. Site-specific: site:[domain.com] — what they publish themselves 7. Companies House / SEC: "[Company Name]" "companies house" OR "SEC filing" — official records

Phase 2: Deep Dive

Based on Phase 1 results, fetch the most promising URLs:

Company About page and product pages
Crunchbase or PitchBook profile
Recent blog posts or press releases (last 6 months)
Companies House / SEC filings (if applicable)
G2, Trustpilot, or Capterra review pages
LinkedIn company page

Phase 3: Synthesis

Compile findings into the structured report format below.

Output Contract

Return a SkillResult with:

data.report: The full formatted markdown report (see Report Format below)
data.company_overview: Structured object with name, founded, headquarters, company_type, industry, website, employees
data.leadership: Array of { name, role, background }
data.products: Array of { name, domain, description, target_customer }
data.timeline: Array of { year, milestone }
data.market_position: Object with performance_claims, scale, notable_clients, awards
data.financials: Object with funding_rounds, total_raised, investors, valuation, revenue_signals
data.reputation: Object with platforms array of { name, rating, review_count, summary }
data.recent_activity: Array of { date, event, source_url }
data.competitive_landscape: Object with direct_competitors, differentiators, market_trends
data.buying_signals: Array of detected buying intent signals (see Buying Signals section)
data.sources: Array of { title, url } for all consulted sources
references: Array of all source URLs used

Report Format

The report field must use this exact markdown structure. Omit sections where no data was found rather than guessing.

## [Company Name] — Intelligence Report

### Company Overview

| Field | Detail |
|-------|--------|
| **Founded** | [Year], [Location] |
| **Headquarters** | [City, Country] |
| **Company Type** | [Private/Public, Company Number if available] |
| **Industry** | [Primary industry] |
| **Website** | [Primary domain] |
| **Employees** | [Estimate or range] |

[2-3 sentence summary of what the company does and its market position.]

---

### Leadership

| Person | Role | Background |
|--------|------|------------|
| [Name] | [Title] | [Brief bio] |

---

### Product Suite

- **[Product Name]** ([domain]) — [Description, key features, target customer]

---

### Company Evolution

| Year | Milestone |
|------|-----------|
| [Year] | [Key event] |

---

### Market Position

- **Performance Claims**: [Stated metrics]
- **Scale**: [Users, customers, countries, revenue]
- **Notable Clients**: [Named customers]
- **Awards/Recognition**: [Awards, rankings]

---

### Financial & Funding

| Round | Date | Amount | Investors |
|-------|------|--------|-----------|
| [Type] | [Date] | [Amount] | [Investors] |

---

### Customer Reputation

| Platform | Rating | Reviews | Summary |
|----------|--------|---------|---------|
| [Platform] | [Score] | [Count] | [Themes] |

---

### Recent Activity

- [Date]: [Event] ([Source])

---

### Competitive Landscape

- **Direct Competitors**: [List]
- **Differentiators**: [What sets them apart]
- **Market Trends**: [Relevant trends]

---

### Buying Intent Signals

- [Signal type]: [Detail and what it means for outreach]

---

### Sources

- [Source Title](URL)

Buying Intent Signals

Flag these if found -- they're high value for sales teams. Use the products, competitors, and industry from the Organization Context above to identify signals relevant to ${company_name}'s sales motion:

Leadership changes: New CTO/VP Engineering = potential tech stack review
Funding round: Fresh capital = budget for new tools
Job postings: Hiring for roles related to ${company_name}'s product area (as described in Organization Context) = active need
Competitor mentions: Complaints about current vendor = switching intent
Conference attendance: Speaking/sponsoring = active in market
Blog content: Writing about problems that ${company_name}'s products solve (reference Organization Context for product details) = awareness stage
Tech stack changes: Migrating platforms = evaluation window
Expansion signals: New offices, new markets, international growth

Quality Standards

Every claim needs a source. Prefix uncertain information with "reportedly" or "according to [source]".
Dates matter. Note when information was published. Flag data older than 6 months.
No hallucinated metrics. If data isn't publicly available, say so explicitly.
Actionable for sales. Include details a rep would use: tech stack, pain points, growth trajectory, decision-makers, buying signals.
Sources section is mandatory. List all URLs with descriptive titles.

Edge Cases

Company not found

Try alternate names, abbreviations, parent companies, domain variations. If still minimal: report what was found and clearly state the gaps.

Multiple companies with same name

Use founder_name or industry input to disambiguate. If still ambiguous, present the most likely match and note the disambiguation.

Very large companies (Fortune 500)

Focus on the most recent 12 months. Skip exhaustive history. Emphasize recent strategic moves, leadership changes, and market positioning shifts.

Very small / new companies

Lean harder on social media, founder backgrounds, job postings, and product screenshots. Note limited data availability honestly.

Error Handling

If company name is ambiguous, use additional context (industry, founder, domain) to disambiguate
If web search returns conflicting information, present the most recent and most credible source
Always return at least company_overview and report even if other sections have limited data
Mark confidence level on uncertain information: "confirmed", "estimated", "unverified"