company-research
starsforks
How to install company-research?
xuxchloris/export_skills/company-research
$npx -y skills add Xuxchloris/export_skills --skill company-researchUse when a trade agent needs to analyze a prospect company, website, business type, product line, purchasing possibility, red flags, decision-maker clues, or personalization evidence
SKILL.md
| 1 | # Company Research |
| 2 | |
| 3 | ## Overview |
| 4 | |
| 5 | Turn prospect website content into evidence-backed sales intelligence. The core rule is to separate observed facts from inferred opportunities. |
| 6 | |
| 7 | ## When to Use |
| 8 | |
| 9 | Use this skill when the user provides a company name, website URL, prospect list row, LinkedIn snippet, trade-show exhibitor page, or distributor page and asks whether the company is worth developing. |
| 10 | |
| 11 | Do not use it to fabricate contact names or emails. |
| 12 | |
| 13 | ## Inputs |
| 14 | |
| 15 | - Company name |
| 16 | - Website URL or pasted website content |
| 17 | - Tool output from `tools/batch_prospect_pipeline.py` or `tools/decision_maker_finder.py` when a URL/list is provided |
| 18 | - Product context from `product-loader` |
| 19 | - Market context from `MARKET.yaml` |
| 20 | |
| 21 | ## Outputs |
| 22 | |
| 23 | ```json |
| 24 | { |
| 25 | "company_summary": "", |
| 26 | "business_type": "", |
| 27 | "main_products": [], |
| 28 | "target_customers": "", |
| 29 | "countries_served": [], |
| 30 | "evidence": [], |
| 31 | "evidence_status": "verified|no_evidence|fetch_failed", |
| 32 | "possible_needs": [], |
| 33 | "personalization_points": [], |
| 34 | "decision_maker_clues": [], |
| 35 | "contact_email": "", |
| 36 | "contact_phone": "", |
| 37 | "email_result": "found|没有", |
| 38 | "phone_result": "found|没有", |
| 39 | "red_flags": [], |
| 40 | "confidence": "low|medium|high" |
| 41 | } |
| 42 | ``` |
| 43 | |
| 44 | ## Procedure |
| 45 | |
| 46 | 1. If the user provides a URL or list file and no fetched content, run `tools/batch_prospect_pipeline.py` or the relevant crawler-backed tool first. Required command reference: run tools/batch_prospect_pipeline.py before scoring or email drafting. |
| 47 | 2. Read the provided website content, browsing result, or `research_reports.json` output. |
| 48 | 3. Do not output company facts without fetched evidence. If fetching fails, return `evidence_status: "fetch_failed"` and stop at review notes. |
| 49 | 4. Extract observed facts: products, business type, brands, market focus, contact page, about page, and catalog clues. |
| 50 | 5. If the input is a company name or company link, resolve the official website when possible and run official website contact search across homepage, contact, about, team, catalog, and product pages. |
| 51 | 6. Record visible email and phone values as `contact_email` and `contact_phone`; if either value is not found, write `email_result: "没有"` or `phone_result: "没有"` instead of guessing. |
| 52 | 7. Record every fact with an evidence URL and short evidence text. |
| 53 | 8. Identify business type using evidence: importer, distributor, wholesaler, retailer, brand owner, manufacturer, contractor, marketplace seller, or unrelated site. |
| 54 | 9. Compare observed facts against the product context. |
| 55 | 10. Create `possible_needs` only when there is visible product, category, application, or channel overlap. |
| 56 | 11. If fetched pages contain no relevant product or channel clue, return `evidence_status: "no_evidence"` and keep confidence low. |
| 57 | 12. Create personalization points from concrete website facts, not generic praise. |
| 58 | 13. Flag red flags: no business relevance, consumer-only content, inactive website, unverifiable contact, unrelated industry, or low-quality scraped directory. |
| 59 | 14. Return confidence based on evidence quantity and recency. |
| 60 | |
| 61 | ## Verification |
| 62 | |
| 63 | - Every personalization point has website evidence. |
| 64 | - Every observed fact includes an evidence URL. |
| 65 | - Official website contact search is run when a company name, website, or company link is available. |
| 66 | - Contact outputs include `contact_email`, `contact_phone`, `email_result`, and `phone_result`. |
| 67 | - Inferences are labeled as possible needs, not facts. |
| 68 | - No contact name is invented. |
| 69 | - Red flags are present when the website is weak or unrelated. |
| 70 | - Confidence is `low` when evidence is thin. |
| 71 | |
| 72 | ## Common Mistakes |
| 73 | |
| 74 | | Mistake | Fix | |
| 75 | | --- | --- | |
| 76 | | Calling every distributor a strong lead | Require product or channel overlap | |
| 77 | | Saying the company imports a product with no proof | Use "may need" and cite the clue | |
| 78 | | Writing generic personalization | Quote a concrete category, market, or service | |
| 79 | | Ignoring weak websites | Mark confidence and red flags clearly | |
People who liked this also viewed
6 picksType
- seo-auditWhen the user wants to audit, review, or diagnose SEO issues on their site. Also use when the user mentions "SEO audit," "technical SEO," "why am I not ranking," "SEO issues," "on-page SEO," "meta tagSkillsJul 2026
- copywritingWhen the user wants to write, rewrite, or improve marketing copy for any page — including homepage, landing pages, pricing pages, feature pages, about pages, or product pages. Also use when the user sSkillsJul 2026
- marketing-psychologyWhen the user wants to apply psychological principles, mental models, or behavioral science to marketing. Also use when the user mentions 'psychology,' 'mental models,' 'cognitive bias,' 'persuasion,'SkillsJul 2026
- content-strategyWhen the user wants to plan a content strategy, decide what content to create, or figure out what topics to cover. Also use when the user mentions "content strategy," "what should I write about," "conSkillsJul 2026
- programmatic-seoWhen the user wants to create SEO-driven pages at scale using templates and data. Also use when the user mentions "programmatic SEO," "template pages," "pages at scale," "directory pages," "location pSkillsJul 2026
- marketing-ideasWhen the user needs marketing ideas, inspiration, or strategies for their SaaS or software product. Also use when the user asks for 'marketing ideas,' 'growth ideas,' 'how to market,' 'marketing stratSkillsJul 2026
