| name | company-research |
| description | Company research using Exa search. Finds company context, competitors, and market signals. |
| context | fork |
Company Research (Exa)
Use this skill for company discovery, competitor scans, vendor profiles, funding or hiring signals, market maps, and quick diligence.
Tool Selection
| Intent | Primary Tool | Notes |
|---|
| Broad company or competitor discovery | web_search_exa | Best first pass when terminology, category fit, or target companies are unclear. |
| Company-profile style results | web_search_advanced_exa with category: "company" | Use when available; do not use deep search types here. |
| Comparative write-up across companies | web_research_exa | Use when enabled and the user needs synthesis or recommendations. |
| Direct factual question about one company | web_answer_exa | Good for concise cited answers. |
| Read selected company pages, filings, or posts | web_fetch_exa | Fetch 1-3 high-signal URLs after discovery. |
| Find companies or pages similar to one strong source | web_find_similar_exa | Use when a seed URL is clearly representative. |
If web_search_advanced_exa is unavailable, use web_search_exa with stronger query wording. If web_research_exa is unavailable, stop at discovery plus a concise manual synthesis.
Category Behavior
category: "company" focuses on company-facing pages and profile-like structure.
- Deep search types (
deep-reasoning, deep-lite, deep) belong to web_research_exa, not web_search_advanced_exa.
Filter Restrictions
When using web_search_advanced_exa with category: "company", avoid:
excludeDomains
startPublishedDate / endPublishedDate
These combinations are rejected by the extension before reaching Exa. If you need freshness or exclusions, run broader discovery first and filter locally.
Recommended Settings
- Broad discovery
{ "query": "AI infrastructure startups serving enterprise developers 2025", "numResults": 10 }
- Company category search
{ "query": "AI infrastructure developer tools startups", "category": "company", "type": "auto", "numResults": 12 }
- Comparative synthesis
{ "query": "Compare AI infrastructure startups serving enterprise developer teams, focusing on product positioning, traction signals, and differentiation.", "systemPrompt": "Prefer official sources, filings, funding announcements, customer pages, and reputable news. Avoid overclaiming from sparse data.", "outputSchema": { "type": "text" } }
Output Guidance
- Return a structured findings list first: company, what they do, signals, and links.
- Distinguish direct evidence from interpretation.
- Call out stale, sparse, or conflicting data.
- Avoid implying private traction or revenue unless sourced.