Recipe: Patent due diligence
Patent landscape for a company or technology area: top patents, assignees, citation density, IP risk factors from SEC filings.
Tools used: search_patents, get_patent, edgar_search_filings, trademark_search.
Calls: 3–6.
Sequence
1. Patent search
search_patents({ query: "CRISPR gene editing" })
// or: search_patents({ query: "Tesla autonomous driving" })Returns title, abstract, assignee, filing date, patent number for the top matches.
2. Drill into top patents
get_patent({ number: "US10123456" })Returns full claims, citations, family members, prosecution history.
For multiple patents, parallelize:
const top = await search_patents({ query: "CRISPR gene editing" })
const detailed = await Promise.all(
top.results.slice(0, 5).map(p => get_patent({ number: p.patent_number }))
)3. Map assignees
Group the search results by assignee. The companies with the most patents in your subject are the IP leaders. Cross-reference with EDGAR for public ones:
// For each top assignee that's public:
resolve_entity({ type: "company", value: "Vertex Pharmaceuticals" })
edgar_search_filings({ query: "Vertex CRISPR" })Look for IP-related disclosures in 10-K Item 1A risk factors.
4. Trademarks for product names
trademark_search({ query: "Tesla", limit: 20 })Cross-check trademark portfolio. Active oppositions are a separate signal.
Citation pattern
CRISPR gene editing patent landscape: top assignees by active patent count are Broad Institute, MIT, and Vertex Pharmaceuticals (per USPTO data). Most-cited foundational patent is US 8,697,359 (Doudna et al.). Vertex disclosed CRISPR-related litigation in its 10-K Item 1A.
Caveats
search_patentsis keyword search. It catches mentions, not necessarily core inventions. For freedom-to-operate, you want a patent attorney, not Pipeworx.- Assignee normalization. “Vertex Pharmaceuticals Inc.”, “Vertex Pharmaceuticals”, and “VERTEX PHARMACEUTICALS, INC.” are all the same company. Group case-insensitive after stripping legal suffixes.
- Foreign filings not covered. USPTO only. EPO, JPO, CNIPA need separate sources.
- Trademark vs. patent. Different protections, different timelines. A strong trademark portfolio doesn’t imply patent strength.