Skip to main content
Glama

Epo Ops

by io.github.pipeworx-io
OverviewDiscussionsAdmin

Search Patents

search_patents
Read-onlyIdempotent

Search published patents worldwide via EPO OPS and get back full bibliographic records — title, applicants, inventors, publication date, country, IPC classifications — for each hit. Searches EPO DOCDB, which indexes publications from 100+ patent offices, so Chinese, Japanese, Korean, US and PCT/WO documents are all in scope alongside European ones — PCT international applications (publications beginning "WO") are covered here and in every other tool in this pack, not just EP filings. Each hit's kind code tells you what stage it is: an "A"-prefixed kind (A1, A2, A3…) is a published APPLICATION, not yet granted; a "B"-prefixed kind (B1, B2…) is a GRANTED patent; a WO publication's "A1"/"A2"/"A3" marks the PCT publication itself, not a grant — PCT applications are examined and granted nationally after entering each country's national phase, which this pack does not track. Results are not filtered to grants by default — most hits from a bare keyword search will be "A" kind. To find issued/granted patents specifically, scope the query with a kind or grant-relevant field (e.g. add AND pd>=<date> and check kind on the returned rows) rather than assuming an unfiltered hit is granted. Example queries: "ta=hydrogen", "in=Tesla", "pa=apple", "txt=neural network AND pd>=2020", "pa=moderna and pn=WO" (a company's PCT filings).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryYesEPO CQL — fields: ta (title-abstract), ti (title), ab (abstract), txt (any text), in (inventor), pa (applicant), cl (classification), pn (publication number), ap (application number), pr (priority), pd (publication date), ad (application date).
rangeNoResult range "start-end" (1-indexed). Max 100 per page. Default "1-25".

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryNoQuery metadata
resultNoSearch result set

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • addedInput schema / examples
      Added value: +[
+  {
+    "query": "ta=hydrogen fuel cell"
+  },
+  {
+    "query": "pa=Tesla AND pd>=2020",
+    "range": "1-50"
+  }
+]
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
+  "description": "Search results from EPO patent database",
+  "properties": {
+    "query": {
+      "description": "Query metadata",
+      "properties": {
+        "syntax": {
+          "description": "Query syntax version",
+          "type": "string"
+        }
+      },
+      "type": "object"
+    },
+    "result": {
+      "description": "Search result set",
+      "properties": {
+        "patents": {
+          "description": "List of patent publications in result",
+          "items": {
+            "properties": {
+              "id": {
+                "description": "Patent ID",
+                "type": "integer"
+              },
+              "publication-reference": {
+                "description": "Publication reference details",
+                "type": "object"
+              }
+            },
+            "type": "object"
+          },
+          "type": "array"
+        },
+        "query-string": {
+          "description": "Original query string",
+          "type": "string"
+        },
+        "total-patent-count": {
+          "description": "Total number of matching patents",
+          "type": "integer"
+        }
+      },
+      "type": "object"
+    }
+  },
+  "type": "object"
+}
  2. First observed

TDQS

A4.6/5.0
Behavior5/5

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

With annotations already covering the safety profile (readOnly, idempotent, openWorld, non-destructive), the description adds substantial behavioral context beyond them: results are NOT filtered to grants by default, most bare keyword hits will be 'A' kind, kind codes encode application vs grant stage, and PCT national-phase grants are not tracked. This materially changes how an agent interprets results.

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

Conciseness4/5

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

The purpose and example queries are front-loaded and every major sentence carries real value for a complex domain. However, the PCT/'WO' caveat is restated several times (coverage in every tool, the A1/A2/A3 marking, national-phase non-tracking), which is more repetition than a 2-parameter tool needs.

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?

Given an output schema exists, return values need not be spelled out, yet the description still frames the record shape and, more importantly, supplies the domain knowledge (kind codes, grant vs application, DOCDB coverage) an agent needs to query and interpret correctly. Nothing essential is missing.

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%, so both parameters are documented, setting a baseline of 3. The description goes beyond the schema by demonstrating CQL composition in practice ('txt=neural network AND pd>=2020', 'pa=moderna and pn=WO') and clarifying the kind/pd scoping pattern, which the schema's field list alone does not convey.

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 published patents worldwide via EPO OPS') and names the exact return payload ('full bibliographic records — title, applicants, inventors, publication date, country, IPC classifications'). This is clearly distinguished from the sibling get_biblio/get_claims/get_abstract tools, which fetch details for a known document rather than finding documents.

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 clear context for use plus explicit filtering advice ('To find issued/granted patents specifically, scope the query with a kind or grant-relevant field') and five worked example queries. It does not explicitly name alternatives in the pack (e.g. get_claims for claim text, search_within for narrowing), so no exclusions/routing to siblings.

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

All Epo Ops toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources