Skip to main content
Glama

Patents Search Patents

patents_search_patents
Read-onlyIdempotent

Search USPTO patent applications and grants, without a granted_after/granted_before bound this returns the APPLICATION corpus sorted newest-FILED-first, not issued patents. USPTO publishes an application ~18 months after filing and typically takes 2+ years to grant it, so the newest rows in an unfiltered search are always recent, unexamined filings with grant_date: null, that is expected, not stale data. Pick the date field on the QUESTION'S VERB, not on the word "recent": a question asking who has FILED / applied for / sought patents in a window wants filed_after/filed_before, the default application corpus already includes pending AND granted patents, so no granted_after is needed or wanted, and adding one will silently drop recently-filed-but-not-yet-examined applications (grant lags filing by 2+ years, so "filed in the last 3 years" AND "granted in the last 3 years" are mostly DISJOINT sets). Only use granted_after/granted_before when the question explicitly says GRANTED / ISSUED / APPROVED / "has a patent" (not just "recent patents"). Use query for free-text keywords ("lithium battery", "crispr", "machine learning"); all terms are required (AND), and you can quote a phrase to keep it together. A curated, short list of known synonym/abbreviation pairs for concepts that patents describe inconsistently (e.g. cfRNA / cell-free RNA / circulating RNA; MRD / minimal residual disease / residual cancer) is recognized in query, quoted or bare, and expanded to an OR group before the search runs, so "cfRNA" and "cell-free RNA" are treated as the same concept rather than requiring the caller to guess a specific patent's exact wording (marked synonyms_expanded: true in the response; this is a fixed, documented list, not general synonym invention). A question that names several DISTINCT ways of describing the same idea (e.g. a condition plus a detection method, each expressed as its own quoted phrase) should still be passed as one query, if the literal AND-of-everything match returns zero, the tool automatically retries once with those same quoted phrases OR'd together instead and marks the response broadened: true; it only ever relaxes multi-phrase AND to OR on a genuine zero, so a search that is narrow for a real reason (e.g. one specific applicant with no matching filings, or an ordinary bare-keyword AND) still correctly returns zero. Optional structured filters: applicant (exact corporate name as filed, e.g. "APPLE INC."), inventor (person name), title (words in the invention title), number (a specific application number), filed_after / filed_before, granted_after / granted_before. Common synonyms are understood, assignee, company and owner all reach applicant, and keywords, q or text all reach query. Results include title, application number, filing date, first applicant, all applicants, inventors, status, classification. total is the full match count but USPTO returns at most 25 records per search, narrow with applicant or a date range rather than raising limit. Powered by the USPTO Open Data Portal (data.uspto.gov).

Example tool call

This call uses example arguments supplied in the tool’s input schema.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "query": "\"cell-free RNA\"",
      "filed_after": "2023-09-25",
      "filed_before": "2026-09-25"
    },
    "name": "patents_search_patents"
  }
}

Input Schema

NameRequiredDescriptionDefault
limitNoNumber of results to return (default 10). USPTO caps every search at 25 records, so values above 25 have no effect, use the filters to narrow instead.
queryNoFree-text keywords. Every term must appear (they are AND-ed), so add words to narrow and remove words to widen. Wrap words in double quotes to require them adjacent: `"machine learning" model` needs the exact phrase plus the word model. Examples: "lithium battery", "crispr", "neural network". Pass "*" if you only want to filter by applicant/date with no keyword constraint. A curated set of known synonyms/abbreviations (e.g. cfRNA / cell-free RNA / circulating RNA, or MRD / minimal residual disease / residual cancer) is automatically expanded to an OR group within the query, quoted or bare, either form is recognized, so you do not need to guess which exact wording a given patent uses (see `synonyms_expanded` in the response). If several quoted phrases, or two recognized synonym-group concepts, quoted or not, ANDed together still match nothing, the tool retries once with them OR'd instead (see `broadened` in the response) rather than returning a bare zero.
titleNoOptional. Words that must appear in the invention title, which narrows far harder than `query` does since `query` searches the whole record. Example: "solid state battery".
numberNoOptional. A specific US application number, digits only or formatted, "16123456" or "16/123,456". Accepted synonyms: `application_number`, `patent_number`.
_apiKeyNoUSPTO ODP API key. Get free at https://data.uspto.gov/myodp. Falls back to platform key if configured.
inventorNoOptional. Inventor name, matched against ALL named inventors (not just the first); "Geoffrey Hinton" or "Hinton, Geoffrey" matches "Geoffrey E. Hinton" (middle initials tolerated), a bare surname matches any inventor with it. Each row carries `matching_inventor`, the inventor who matched. Examples: "Geoffrey Hinton", "Bengio". Accepted synonyms: `inventor_name`, `author`.
applicantNoOptional. Company applicant name as it appears on the USPTO filing. This is also how to list the patents a company OWNS or holds ("patents assigned to X", "X's patents", "patents owned by X"): set `applicant`, with `query` "*" if there is no keyword. **Must include the exact corporate suffix** the company uses (PBC / Inc. / LLC / Corporation / Co. / NV / AG / KK). A wrong or missing suffix matches nothing, "Apple" returns zero where "APPLE INC." returns hundreds. Examples: "Anthropic, PBC" (not "Anthropic Inc."), "Apple Inc." (not "Apple"), "Alphabet Inc." (not "Google"), "Meta Platforms, Inc." (not "Facebook"), "Microsoft Corporation" (not "Microsoft Corp."). If you get zero results plus a `warning` field, the name form is wrong rather than the company being absent, retry with a different corporate form.
filed_afterNoOptional. Filter to patents FILED on/after this date (ISO YYYY-MM-DD). Use this for "who has filed / applied for patents on X (recently / in the last N years)", it covers both pending applications and already-granted patents, so it is the right bound even when the question also says "recent". Do NOT substitute `granted_after` for a filing-activity question: grant lags filing by 2+ years, so a recent filing window under `granted_after` typically returns zero even when filing activity is real.
filed_beforeNoOptional. Filter to patents FILED on/before this date (ISO YYYY-MM-DD).
granted_afterNoOptional. Filter to patents GRANTED (issued) on/after this date (ISO YYYY-MM-DD), use ONLY when the question explicitly says GRANTED / ISSUED / APPROVED, not for a plain "recent patents" or "who has filed" question (use `filed_after` for those, see its description). Accepted synonym: `issued_after`.
granted_beforeNoOptional. Filter to patents GRANTED (issued) on/before this date (ISO YYYY-MM-DD). Accepted synonym: `issued_before`.

Output Schema

NameRequiredDescriptionDefault
noteNoPresent when total exceeds the returned records, explains the 25-record ODP page cap
queryYesThe composed ODP query string actually sent
totalYesTotal number of matching applications (USPTO ODP returns at most 25 records per search regardless of limit)
filtersYesEcho of the structured filters applied; each is null when unused
patentsYesSame records with full ODP fields
resultsYesBack-compat summary shape (title, number, dates, applicant)
warningNoPresent when an applicant filter matched nothing, ODP matches the corporate name literally ("APPLE INC." not "Apple")
returnedYesNumber of records in this response

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedOutput schema / properties / note / description
      Previous value: -"Present when total exceeds the returned records — explains the 25-record ODP page cap"New value: +"Present when total exceeds the returned records, explains the 25-record ODP page cap"
    • changedOutput schema / properties / warning / description
      Previous value: -"Present when an applicant filter matched nothing — ODP matches the corporate name literally (\"APPLE INC.\" not \"Apple\")"New value: +"Present when an applicant filter matched nothing, ODP matches the corporate name literally (\"APPLE INC.\" not \"Apple\")"
  2. Changed6 schema fields changed
    • changedInput schema / properties / applicant / description
      Previous value: -"Optional. Company applicant name as it appears on the USPTO filing. This is also how to list the patents a company OWNS or holds (\"patents assigned to X\", \"X's patents\", \"patents owned by X\"): set `applicant`, with `query` \"*\" if there is no keyword. **Must include the exact corporate suffix** the company uses (PBC / Inc. / LLC / Corporation / Co. / NV / AG / KK). A wrong or missing suffix matches nothing — \"Apple\" returns zero where \"APPLE INC.\" returns hundreds. Examples: \"Anthropic, PBC\" (not \"Anthropic Inc.\"), \"Apple Inc.\" (not \"Apple\"), \"Alphabet Inc.\" (not \"Google\"), \"Meta Platforms, Inc.\" (not \"Facebook\"), \"Microsoft Corporation\" (not \"Microsoft Corp.\"). If you get zero results plus a `warning` field, the name form is wrong rather than the company being absent — retry with a different corporate form."New value: +"Optional. Company applicant name as it appears on the USPTO filing. This is also how to list the patents a company OWNS or holds (\"patents assigned to X\", \"X's patents\", \"patents owned by X\"): set `applicant`, with `query` \"*\" if there is no keyword. **Must include the exact corporate suffix** the company uses (PBC / Inc. / LLC / Corporation / Co. / NV / AG / KK). A wrong or missing suffix matches nothing, \"Apple\" returns zero where \"APPLE INC.\" returns hundreds. Examples: \"Anthropic, PBC\" (not \"Anthropic Inc.\"), \"Apple Inc.\" (not \"Apple\"), \"Alphabet Inc.\" (not \"Google\"), \"Meta Platforms, Inc.\" (not \"Facebook\"), \"Microsoft Corporation\" (not \"Microsoft Corp.\"). If you get zero results plus a `warning` field, the name form is wrong rather than the company being absent, retry with a different corporate form."
    • changedInput schema / properties / filed_after / description
      Previous value: -"Optional. Filter to patents FILED on/after this date (ISO YYYY-MM-DD). Use this for \"who has filed / applied for patents on X (recently / in the last N years)\" — it covers both pending applications and already-granted patents, so it is the right bound even when the question also says \"recent\". Do NOT substitute `granted_after` for a filing-activity question: grant lags filing by 2+ years, so a recent filing window under `granted_after` typically returns zero even when filing activity is real."New value: +"Optional. Filter to patents FILED on/after this date (ISO YYYY-MM-DD). Use this for \"who has filed / applied for patents on X (recently / in the last N years)\", it covers both pending applications and already-granted patents, so it is the right bound even when the question also says \"recent\". Do NOT substitute `granted_after` for a filing-activity question: grant lags filing by 2+ years, so a recent filing window under `granted_after` typically returns zero even when filing activity is real."
    • changedInput schema / properties / granted_after / description
      Previous value: -"Optional. Filter to patents GRANTED (issued) on/after this date (ISO YYYY-MM-DD) — use ONLY when the question explicitly says GRANTED / ISSUED / APPROVED, not for a plain \"recent patents\" or \"who has filed\" question (use `filed_after` for those — see its description). Accepted synonym: `issued_after`."New value: +"Optional. Filter to patents GRANTED (issued) on/after this date (ISO YYYY-MM-DD), use ONLY when the question explicitly says GRANTED / ISSUED / APPROVED, not for a plain \"recent patents\" or \"who has filed\" question (use `filed_after` for those, see its description). Accepted synonym: `issued_after`."
    • changedInput schema / properties / limit / description
      Previous value: -"Number of results to return (default 10). USPTO caps every search at 25 records, so values above 25 have no effect — use the filters to narrow instead."New value: +"Number of results to return (default 10). USPTO caps every search at 25 records, so values above 25 have no effect, use the filters to narrow instead."
    • changedInput schema / properties / number / description
      Previous value: -"Optional. A specific US application number, digits only or formatted — \"16123456\" or \"16/123,456\". Accepted synonyms: `application_number`, `patent_number`."New value: +"Optional. A specific US application number, digits only or formatted, \"16123456\" or \"16/123,456\". Accepted synonyms: `application_number`, `patent_number`."
    • changedInput schema / properties / query / description
      Previous value: -"Free-text keywords. Every term must appear (they are AND-ed), so add words to narrow and remove words to widen. Wrap words in double quotes to require them adjacent: `\"machine learning\" model` needs the exact phrase plus the word model. Examples: \"lithium battery\", \"crispr\", \"neural network\". Pass \"*\" if you only want to filter by applicant/date with no keyword constraint. A curated set of known synonyms/abbreviations (e.g. cfRNA / cell-free RNA / circulating RNA, or MRD / minimal residual disease / residual cancer) is automatically expanded to an OR group within the query — quoted or bare, either form is recognized — so you do not need to guess which exact wording a given patent uses (see `synonyms_expanded` in the response). If several quoted phrases — or two recognized synonym-group concepts, quoted or not — ANDed together still match nothing, the tool retries once with them OR'd instead (see `broadened` in the response) rather than returning a bare zero."New value: +"Free-text keywords. Every term must appear (they are AND-ed), so add words to narrow and remove words to widen. Wrap words in double quotes to require them adjacent: `\"machine learning\" model` needs the exact phrase plus the word model. Examples: \"lithium battery\", \"crispr\", \"neural network\". Pass \"*\" if you only want to filter by applicant/date with no keyword constraint. A curated set of known synonyms/abbreviations (e.g. cfRNA / cell-free RNA / circulating RNA, or MRD / minimal residual disease / residual cancer) is automatically expanded to an OR group within the query, quoted or bare, either form is recognized, so you do not need to guess which exact wording a given patent uses (see `synonyms_expanded` in the response). If several quoted phrases, or two recognized synonym-group concepts, quoted or not, ANDed together still match nothing, the tool retries once with them OR'd instead (see `broadened` in the response) rather than returning a bare zero."
  3. Changed2 schema fields changed
    • changedOutput schema / properties / patents / items / properties / patent_number / type
      Previous value: -"string"New value: +[
      +  "string",
      +  "null"
      +]
    • changedOutput schema / properties / patents / items / required
      Previous value: -[
      -  "patent_number",
      -  "title",
      -  "date",
      -  "inventors",
      -  "assignee_organization"
      -]New value: +[
      +  "title"
      +]
  4. Added

TDQS

A4.5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare read-only, idempotent, open-world behavior, and the description adds the things annotations cannot: the 25-record USPTO cap, grant-lag semantics, the automatic one-shot AND→OR broadening on zero results, curated synonym expansion, the `warning` signal on a bad corporate suffix, and `total` vs returned rows. Nothing here contradicts the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The critical caveat is front-loaded, but the filed-vs-granted point is repeated three or four times across the description and in the schema, and the synonym/broadening mechanics are explained twice. It is dense rather than wasteful, yet noticeably longer than needed for the information conveyed.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 11-parameter, zero-required search tool with an output schema and full annotations, the description covers corpus semantics, cap behavior, retry behavior, and response markers. An agent has everything needed to call it and interpret the result correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% and each parameter already documents itself, so the baseline is 3. The description still adds decision-level synthesis: choose the date field on the question's verb, use `query` "*" with `applicant` for ownership questions, and exact corporate suffixes are mandatory. It largely restates the schema, which is why it is not a 5.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Search USPTO patent applications and grants') and immediately disambiguates the corpus: an unbounded search returns newest-FILED applications, not issued patents. An agent can predict what comes back without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives unusually explicit when/when-not guidance for the risky choice (filed_* vs granted_*), keyed to the question's verb, including the failure mode of adding granted_after to a filing-activity question. It never routes to sibling tools (get_patent, search_inventors, search_patent_assignments), so tool-level alternatives are left to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.