Search Prior Art
Conduct a structured prior art search to find publications, patents, products, or disclosures that predate a specific invention. Used to assess patentability (can this be patented?), challenge validity (should this patent have been granted?), or establish freedom-to-operate (is this design covered by existing rights?).
When to Use
- Evaluating whether an invention is novel and non-obvious before filing a patent application
- Challenging the validity of an existing patent by finding prior art the examiner missed
- Supporting a freedom-to-operate analysis by finding prior art that limits a blocking patent's scope
- Documenting a defensive publication to prevent others from patenting a concept
- Responding to a patent office action that questions novelty or obviousness
Inputs
- Required: Invention description (what it does, how it works, what problem it solves)
- Required: Search purpose (patentability, invalidity, FTO, defensive)
- Required: Critical date (filing date of the patent application, or invention date for prior art)
- Optional: Known related patents or publications
- Optional: Technology classification codes (IPC, CPC)
- Optional: Key inventors or companies in the field
Procedure
Step 1: Decompose the Invention into Searchable Elements
Break the invention into its constituent technical features.
- Read the invention description (or patent claims if searching against an existing patent)
- Extract the essential elements — each independent technical feature:
- What components does it have?
- What steps does the process follow?
- What technical effect does it achieve?
- What problem does it solve and how?
- Identify the novel combination — what makes this different from the known art:
- Is it a new element added to known elements?
- Is it a new combination of known elements?
- Is it a known element applied in a new field?
- Generate search terms for each element:
- Technical terms, synonyms, and abbreviations
- Broader and narrower terms (hierarchy)
- Alternative descriptions of the same concept
- Document the Search Map: elements, terms, and relationships
Search Map Example:
+------------------+-----------------------------------+-----------+
| Element | Search Terms | Priority |
+------------------+-----------------------------------+-----------+
| Attention layer | attention mechanism, self- | High |
| | attention, multi-head attention | |
| Sparse routing | mixture of experts, sparse MoE, | High |
| | top-k routing, expert selection | |
| Training method | knowledge distillation, teacher- | Medium |
| | student, progressive training | |
+------------------+-----------------------------------+-----------+
Expected: A complete decomposition with search terms for each element. The novel combination is identified — this is what the search must either find (to invalidate) or confirm is absent (to support novelty).
On failure: If the invention is too abstract to decompose, ask for a more specific description. If the claims are unclear, focus on the broadest reasonable interpretation of each claim element.
Step 2: Search Patent Literature
Search patent databases systematically.
- Construct queries combining element terms:
- Search each element individually first (broad)
- Then combine elements to find closer art (narrow)
- Use classification codes to filter by technology area
- Search multiple databases:
- Google Patents: Good for full-text search, free, large corpus
- USPTO PatFT/AppFT: US patents and applications, official source
- Espacenet: European patents, excellent classification search
- WIPO Patentscope: PCT applications, global coverage
- Apply date filters:
- Prior art must predate the critical date (filing date or priority date)
- Include publications up to 1 year before filing (grace period varies by jurisdiction)
- For each relevant result, record:
- Document number, title, filing date, publication date
- Which elements it discloses (map to Search Map)
- Whether it discloses the novel combination
- Classify results by relevance:
- X reference: Discloses the invention alone (anticipation)
- Y reference: Discloses key elements, combinable with other references (obviousness)
- A reference: Background art, defines the general state of the art
Expected: A classified list of patent references mapped to the invention's elements. X references (if found) are showstoppers for novelty. Y references are the building blocks for obviousness arguments.
On failure: If no relevant patent art is found, this doesn't mean the invention is novel — non-patent literature (Step 3) may contain the critical reference. Absence in one database doesn't mean absence everywhere.
Step 3: Search Non-Patent Literature
Search academic papers, products, open source, and other non-patent disclosures.
- Academic literature:
- Google Scholar, arXiv, IEEE Xplore, ACM Digital Library
- Search using the same terms from Step 1
- Conference papers and workshop proceedings often predate patent filings
- Products and commercial disclosures:
- Product documentation, user manuals, marketing materials
- Internet Archive (Wayback Machine) for date-verified web content
- Trade publications and press releases
- Open source and code:
- GitHub, GitLab — search for implementations of the technical features
- README files, documentation, and commit histories for date evidence
- Software releases with version dates
- Standards and specifications:
- IEEE, IETF (RFCs), W3C, ISO standards
- Standards-essential patents must be disclosed; search standard bodies' IP databases
- Defensive publications:
- IBM Technical Disclosure Bulletin
- Research Disclosure journal
- IP.com Prior Art Database
- For each result, verify the publication date is before the critical date:
- Web pages: use Wayback Machine for date evidence
- Software: use release dates or commit timestamps
- Papers: use publication date, not submission date
Expected: Non-patent references that complement the patent search. Academic papers and open-source code are often the most powerful prior art because they tend to describe technical details more explicitly than patents.
On failure: If non-patent literature is sparse, the technology may be primarily developed in corporate R&D (patent-heavy). Shift emphasis to patent literature and focus on the combination-based obviousness argument.
Step 4: Analyze and Map Results
Evaluate how the collected prior art relates to the invention.
- Create a claim chart mapping prior art to invention elements:
Claim Element vs. Prior Art Matrix:
+------------------+--------+--------+--------+--------+
| Element | Ref #1 | Ref #2 | Ref #3 | Ref #4 |
+------------------+--------+--------+--------+--------+
| Element A | X | X | | X |
| Element B | | X | X | |
| Element C | X | | X | |
| Novel combo A+B+C| | | | |
+------------------+--------+--------+--------+--------+
X = element disclosed in this reference
- Assess novelty: Does any single reference disclose all elements?
- If yes → invention is anticipated (not novel)
- If no → invention may be novel (proceed to obviousness)
- Assess obviousness: Can a small number of references (2-3) be combined to cover all elements?
- Is there motivation to combine? (would a skilled person see a reason to combine these?)
- Do the references teach away from the combination? (suggest it wouldn't work?)
- For FTO searches: Does the prior art narrow the blocking patent's claims?
- Prior art that overlaps with the blocking patent's claims limits their enforceable scope
- Document the analysis clearly with citation to specific passages
Expected: A clear claim chart showing which elements are covered by which references, with an assessment of novelty and obviousness. Each mapping cites specific passages or figures in the references.
On failure: If the claim chart shows gaps (elements not found in any prior art), those gaps represent the potentially novel aspects. Focus follow-up searches on those specific gaps.
Step 5: Document and Deliver
Package the search results for their intended use.
- Write the Prior Art Search Report:
- Purpose and scope of the search
- Search methodology (databases, queries, date ranges)
- Results summary (number of references found, classification breakdown)
- Top references with detailed analysis (claim charts)
- Assessment: novelty, obviousness, and FTO implications
- Limitations and recommendations for further search
- Organize references:
- Sorted by relevance (X references first, then Y, then A)
- Each reference with full bibliographic data and access link
- Key passages highlighted or extracted
- Recommendations based on search purpose:
- Patentability: File/don't file, suggested claim scope based on prior art gaps
- Invalidity: Strongest combination of references, suggested legal argument
- FTO: Risk level, design-around opportunities, licensing considerations
- Defensive: Whether to publish as defensive disclosure based on white space found
Expected: A complete, well-organized search report that directly supports the intended decision. References are accessible and analysis is traceable.
On failure: If the search is inconclusive (no strong X or Y references, but some relevant background), state the conclusion clearly: "No anticipatory art found; closest art addresses elements A and B but not C. Recommend filing with claims emphasizing element C." Inconclusive is a valid and useful result.
Validation Checklist
Common Pitfalls
- Keyword tunnel vision: Searching only exact terms misses synonyms and alternative descriptions. Use the term hierarchy from Step 1
- Patent-only search: Non-patent literature (papers, products, code) is often more explicit than patents. Don't skip Step 3
- Date carelessness: Prior art must predate the critical date. A brilliant reference from one day after the filing date is worthless
- Ignoring foreign language art: Major inventions may first appear in Chinese, Japanese, Korean, or German patent literature. Machine translation makes these searchable
- Confirmation bias: Searching to confirm novelty rather than searching to find invalidating art. The best search tries hardest to find the closest art
- Stopping too early: The first few results are rarely the best. Iterate search terms based on what early results reveal about the field's vocabulary
Related Skills
assess-ip-landscape — Broader landscape mapping that contextualizes specific prior art searches
screen-trademark — Trademark-specific conflict screening (different databases, different legal framework than patent prior art)
file-trademark — Trademark filing procedures for when screening is complete
review-research — Literature review methodology overlaps significantly with prior art search
security-audit-codebase — Systematic search methodology parallels (thoroughness, documentation, reproducibility)
1---2name: search-prior-art3description: >4 Search for prior art relevant to a specific invention or patent claim.5 Covers patent literature, non-patent literature (academic papers, products,6 open source), defensive publications, and standard-essential patents.7 Use when evaluating whether an invention is novel and non-obvious before8 filing, challenging the validity of an existing patent, supporting a9 freedom-to-operate analysis, documenting a defensive publication, or10 responding to a patent office action questioning novelty or obviousness.11license: MIT12allowed-tools: Read Grep Glob WebFetch WebSearch13metadata:14 source: https://skillmd.com/skills/pjt222/search-prior-art15 author: Philipp Thoss16 version: "1.0"17 domain: intellectual-property18 complexity: intermediate19 language: natural20 tags: intellectual-property, prior-art, patents, novelty, obviousness, invalidity, fto21---2223# Search Prior Art2425Conduct a structured prior art search to find publications, patents, products, or disclosures that predate a specific invention. Used to assess patentability (can this be patented?), challenge validity (should this patent have been granted?), or establish freedom-to-operate (is this design covered by existing rights?).2627## When to Use2829- Evaluating whether an invention is novel and non-obvious before filing a patent application30- Challenging the validity of an existing patent by finding prior art the examiner missed31- Supporting a freedom-to-operate analysis by finding prior art that limits a blocking patent's scope32- Documenting a defensive publication to prevent others from patenting a concept33- Responding to a patent office action that questions novelty or obviousness3435## Inputs3637- **Required**: Invention description (what it does, how it works, what problem it solves)38- **Required**: Search purpose (patentability, invalidity, FTO, defensive)39- **Required**: Critical date (filing date of the patent application, or invention date for prior art)40- **Optional**: Known related patents or publications41- **Optional**: Technology classification codes (IPC, CPC)42- **Optional**: Key inventors or companies in the field4344## Procedure4546### Step 1: Decompose the Invention into Searchable Elements4748Break the invention into its constituent technical features.49501. Read the invention description (or patent claims if searching against an existing patent)512. Extract the **essential elements** — each independent technical feature:52 - What components does it have?53 - What steps does the process follow?54 - What technical effect does it achieve?55 - What problem does it solve and how?563. Identify the **novel combination** — what makes this different from the known art:57 - Is it a new element added to known elements?58 - Is it a new combination of known elements?59 - Is it a known element applied in a new field?604. Generate search terms for each element:61 - Technical terms, synonyms, and abbreviations62 - Broader and narrower terms (hierarchy)63 - Alternative descriptions of the same concept645. Document the **Search Map**: elements, terms, and relationships6566```text67Search Map Example:68+------------------+-----------------------------------+-----------+69| Element | Search Terms | Priority |70+------------------+-----------------------------------+-----------+71| Attention layer | attention mechanism, self- | High |72| | attention, multi-head attention | |73| Sparse routing | mixture of experts, sparse MoE, | High |74| | top-k routing, expert selection | |75| Training method | knowledge distillation, teacher- | Medium |76| | student, progressive training | |77+------------------+-----------------------------------+-----------+78```7980**Expected:** A complete decomposition with search terms for each element. The novel combination is identified — this is what the search must either find (to invalidate) or confirm is absent (to support novelty).8182**On failure:** If the invention is too abstract to decompose, ask for a more specific description. If the claims are unclear, focus on the broadest reasonable interpretation of each claim element.8384### Step 2: Search Patent Literature8586Search patent databases systematically.87881. Construct queries combining element terms:89 - Search each element individually first (broad)90 - Then combine elements to find closer art (narrow)91 - Use classification codes to filter by technology area922. Search multiple databases:93 - **Google Patents**: Good for full-text search, free, large corpus94 - **USPTO PatFT/AppFT**: US patents and applications, official source95 - **Espacenet**: European patents, excellent classification search96 - **WIPO Patentscope**: PCT applications, global coverage973. Apply date filters:98 - Prior art must predate the **critical date** (filing date or priority date)99 - Include publications up to 1 year before filing (grace period varies by jurisdiction)1004. For each relevant result, record:101 - Document number, title, filing date, publication date102 - Which elements it discloses (map to Search Map)103 - Whether it discloses the novel combination1045. Classify results by relevance:105 - **X reference**: Discloses the invention alone (anticipation)106 - **Y reference**: Discloses key elements, combinable with other references (obviousness)107 - **A reference**: Background art, defines the general state of the art108109**Expected:** A classified list of patent references mapped to the invention's elements. X references (if found) are showstoppers for novelty. Y references are the building blocks for obviousness arguments.110111**On failure:** If no relevant patent art is found, this doesn't mean the invention is novel — non-patent literature (Step 3) may contain the critical reference. Absence in one database doesn't mean absence everywhere.112113### Step 3: Search Non-Patent Literature114115Search academic papers, products, open source, and other non-patent disclosures.1161171. **Academic literature**:118 - Google Scholar, arXiv, IEEE Xplore, ACM Digital Library119 - Search using the same terms from Step 1120 - Conference papers and workshop proceedings often predate patent filings1212. **Products and commercial disclosures**:122 - Product documentation, user manuals, marketing materials123 - Internet Archive (Wayback Machine) for date-verified web content124 - Trade publications and press releases1253. **Open source and code**:126 - GitHub, GitLab — search for implementations of the technical features127 - README files, documentation, and commit histories for date evidence128 - Software releases with version dates1294. **Standards and specifications**:130 - IEEE, IETF (RFCs), W3C, ISO standards131 - Standards-essential patents must be disclosed; search standard bodies' IP databases1325. **Defensive publications**:133 - IBM Technical Disclosure Bulletin134 - Research Disclosure journal135 - IP.com Prior Art Database1366. For each result, verify the **publication date** is before the critical date:137 - Web pages: use Wayback Machine for date evidence138 - Software: use release dates or commit timestamps139 - Papers: use publication date, not submission date140141**Expected:** Non-patent references that complement the patent search. Academic papers and open-source code are often the most powerful prior art because they tend to describe technical details more explicitly than patents.142143**On failure:** If non-patent literature is sparse, the technology may be primarily developed in corporate R&D (patent-heavy). Shift emphasis to patent literature and focus on the combination-based obviousness argument.144145### Step 4: Analyze and Map Results146147Evaluate how the collected prior art relates to the invention.1481491. Create a **claim chart** mapping prior art to invention elements:150151```text152Claim Element vs. Prior Art Matrix:153+------------------+--------+--------+--------+--------+154| Element | Ref #1 | Ref #2 | Ref #3 | Ref #4 |155+------------------+--------+--------+--------+--------+156| Element A | X | X | | X |157| Element B | | X | X | |158| Element C | X | | X | |159| Novel combo A+B+C| | | | |160+------------------+--------+--------+--------+--------+161X = element disclosed in this reference162```1631642. Assess **novelty**: Does any single reference disclose all elements?165 - If yes → invention is anticipated (not novel)166 - If no → invention may be novel (proceed to obviousness)1673. Assess **obviousness**: Can a small number of references (2-3) be combined to cover all elements?168 - Is there motivation to combine? (would a skilled person see a reason to combine these?)169 - Do the references teach away from the combination? (suggest it wouldn't work?)1704. For **FTO searches**: Does the prior art narrow the blocking patent's claims?171 - Prior art that overlaps with the blocking patent's claims limits their enforceable scope1725. Document the analysis clearly with citation to specific passages173174**Expected:** A clear claim chart showing which elements are covered by which references, with an assessment of novelty and obviousness. Each mapping cites specific passages or figures in the references.175176**On failure:** If the claim chart shows gaps (elements not found in any prior art), those gaps represent the potentially novel aspects. Focus follow-up searches on those specific gaps.177178### Step 5: Document and Deliver179180Package the search results for their intended use.1811821. Write the **Prior Art Search Report**:183 - Purpose and scope of the search184 - Search methodology (databases, queries, date ranges)185 - Results summary (number of references found, classification breakdown)186 - Top references with detailed analysis (claim charts)187 - Assessment: novelty, obviousness, and FTO implications188 - Limitations and recommendations for further search1892. Organize references:190 - Sorted by relevance (X references first, then Y, then A)191 - Each reference with full bibliographic data and access link192 - Key passages highlighted or extracted1933. Recommendations based on search purpose:194 - **Patentability**: File/don't file, suggested claim scope based on prior art gaps195 - **Invalidity**: Strongest combination of references, suggested legal argument196 - **FTO**: Risk level, design-around opportunities, licensing considerations197 - **Defensive**: Whether to publish as defensive disclosure based on white space found198199**Expected:** A complete, well-organized search report that directly supports the intended decision. References are accessible and analysis is traceable.200201**On failure:** If the search is inconclusive (no strong X or Y references, but some relevant background), state the conclusion clearly: "No anticipatory art found; closest art addresses elements A and B but not C. Recommend filing with claims emphasizing element C." Inconclusive is a valid and useful result.202203## Validation Checklist204205- [ ] Invention decomposed into distinct searchable elements206- [ ] Novel combination explicitly identified207- [ ] Patent databases searched (minimum 2 databases)208- [ ] Non-patent literature searched (academic + products + open source)209- [ ] All references predate the critical date (dates verified)210- [ ] Claim chart maps elements to references with passage citations211- [ ] Novelty and obviousness assessed with reasoning212- [ ] Results classified by relevance (X, Y, A references)213- [ ] Report includes methodology, limitations, and recommendations214- [ ] Search is reproducible (queries and databases documented)215216## Common Pitfalls217218- **Keyword tunnel vision**: Searching only exact terms misses synonyms and alternative descriptions. Use the term hierarchy from Step 1219- **Patent-only search**: Non-patent literature (papers, products, code) is often more explicit than patents. Don't skip Step 3220- **Date carelessness**: Prior art must predate the critical date. A brilliant reference from one day after the filing date is worthless221- **Ignoring foreign language art**: Major inventions may first appear in Chinese, Japanese, Korean, or German patent literature. Machine translation makes these searchable222- **Confirmation bias**: Searching to confirm novelty rather than searching to find invalidating art. The best search tries hardest to find the closest art223- **Stopping too early**: The first few results are rarely the best. Iterate search terms based on what early results reveal about the field's vocabulary224225## Related Skills226227- `assess-ip-landscape` — Broader landscape mapping that contextualizes specific prior art searches228- `screen-trademark` — Trademark-specific conflict screening (different databases, different legal framework than patent prior art)229- `file-trademark` — Trademark filing procedures for when screening is complete230- `review-research` — Literature review methodology overlaps significantly with prior art search231- `security-audit-codebase` — Systematic search methodology parallels (thoroughness, documentation, reproducibility)