---
name: patent-search
description: Use this skill when you need a fast, structured prior-art search and patent-landscape scan before deciding to file, invest, or engage a patent attorney. It produces a documented search log and an issue map organised against the statutory elements — never an opinion.
---

# Patent Search & Prior-Art

> **What this is** — a repeatable, AI-assisted working method for doing the structured legwork on a request in patent prior-art research and producing a rigorous first-pass draft and issue map quickly, with a qualified patent attorney or agent kept in the loop.
> **What this is NOT** — **not legal advice, and not a freedom-to-operate (FTO), patentability, validity, or infringement opinion.** Only a registered patent attorney or agent can provide those. Search results are never exhaustive; a professional search and opinion are required before filing, relying, or representing patentability. Every finding or document is a draft that requires professional review before it is relied on, deployed, filed, or represented to anyone. No professional-client relationship is created.

## When to use this
- You have an invention concept and want a fast, documented prior-art scan before committing to attorney time.
- You need a landscape view of who is active in a technology area and where the crowded vs. open regions are.
- You want the closest references organised against the statutory elements as an issue map for an attorney.
- You are preparing an inventor disclosure and want a documented search log to hand to counsel.
- You need to decide whether a concept is worth a professional search and opinion at all.

## Operating principle
AI and disciplined search craft can frame the problem, build classification-aware queries, and surface and organise likely-relevant references fast — but a registered patent attorney or agent decides patentability, scope, and risk. The value is a faster, better-organised starting point and a reproducible search log; it is never the final word and never an opinion.

## Capability 1 — Prior-art search strategy
**Goal.** Find and document the most relevant prior art with a reproducible, defensible search log.
**Inputs.** A clear description of the invention, the problem it solves, known competitors/assignees, and any known references.
**Method.**
1. Frame the invention as problem → solution → distinguishing features; list the technical concepts to search.
2. Map concepts to **CPC** and **IPC** classification codes to search by subject matter, not just keywords.
3. Build keyword queries with synonyms, alternate terminology, and industry jargon; combine with class and assignee/inventor filters.
4. Search patent literature across **Google Patents**, **Espacenet**, **USPTO** search / **PatentsView**, and **WIPO PATENTSCOPE**.
5. Search non-patent literature (academic papers, standards, product manuals, datasheets) — prior art is not limited to patents.
6. Iteratively refine: read top hits, harvest new classes/terms, re-query; record each query, database, date, and result count in a search log.
7. Shortlist the closest references and capture, for each, the relevant passages and figures.
**Output.** A ranked shortlist of closest references with citations, plus a complete, reproducible search log.
**Quality bar (what the professional receives).** Documented queries and databases (so the attorney can verify and extend), explicit statement that the search is not exhaustive, and relevant passages highlighted rather than conclusions asserted.

## Capability 2 — Patent landscape analysis
**Goal.** Show the competitive and white-space picture of a technology area.
**Inputs.** The technology definition, relevant CPC/IPC classes, and a set of retrieved records.
**Method.**
1. Cluster retrieved patents by assignee, classification, priority date, and jurisdiction.
2. Plot filing activity over time to show momentum and recent entrants.
3. Identify key players and their apparent focus areas from their portfolios.
4. Map crowded fields vs. apparent white space, noting that "white space" in a search is not a legal clearance.
5. Trace citation networks to find foundational patents and clusters of related work.
**Output.** A landscape summary: key players, activity trends, crowded/open regions, and foundational references.
**Quality bar (what the professional receives).** Every observation is tied to the underlying records and framed as descriptive, with the caveat that landscape gaps do not establish patentability or freedom to operate.

## Capability 3 — Patentability preliminary assessment
**Goal.** Organise the closest art against the statutory elements as an issue map for an attorney.
**Inputs.** The shortlist from Capability 1 and the invention's claimed features.
**Method.**
1. State the invention's key features as candidate claim elements.
2. For each closest reference, map which elements it appears to disclose and which it does not.
3. Organise against **novelty** (anticipation) and **non-obviousness / inventive step** as an issue map — explicitly not a legal conclusion.
4. Flag subject-matter-eligibility questions where relevant (e.g., abstract-idea / software concerns) for counsel.
5. List open questions and the specific references an attorney should scrutinise first.
**Output.** An element-by-reference issue map with open questions flagged for counsel.
**Quality bar (what the professional receives).** A neutral mapping — no patentability verdict, no scope opinion — with every statutory element and reference clearly attributed.

## Worked example (illustrative)
*Illustrative and hypothetical.* An inventor proposes "a solar-powered irrigation valve that adjusts flow using soil-moisture and weather-forecast data." The method frames three concepts (solar power management, soil-moisture sensing, forecast-driven control), maps them to plausible CPC areas (e.g., agricultural irrigation and control), and runs class + keyword queries on Espacenet, Google Patents, and PATENTSCOPE, logging each. Three close references emerge; an element map shows two disclose moisture-driven control but neither combines forecast data with solar power management in the claimed way. This is written up as an **issue map for the attorney** — novelty and non-obviousness questions raised, not answered — with the search log attached. No patentability or FTO conclusion is offered.

## Guardrails & escalation
- Any decision to file, not file, rely on, or represent patentability must go to a registered patent attorney or agent with a professional search and opinion.
- FTO / infringement questions and validity opinions are out of scope and must not be inferred from these outputs.
- No search is exhaustive; the absence of a reference is not evidence a reference does not exist.
- Uncertainty and open questions are flagged explicitly and every close reference is listed for counsel to examine, rather than resolved here.

## References & sources
- USPTO — Patent Public Search and USPTO resources.
- PatentsView — USPTO-backed patent data and search.
- Espacenet (European Patent Office) — worldwide patent search.
- Google Patents — patent and some non-patent literature search.
- WIPO PATENTSCOPE — international (PCT) patent search.
- CPC (Cooperative Patent Classification) and IPC (International Patent Classification) — classification systems.
- 35 U.S.C. § 101 (patentable subject matter), § 102 (novelty), § 103 (non-obviousness) — named for the issue map only.
- EPC Article 54 (novelty) and Article 56 (inventive step) — named for the issue map only.

---
*Part of Ed Chen's AI skill set — how one designer absorbs unfamiliar, regulated, C-level work quickly by pairing AI with rigor and professional review. https://edwson.com*