duncanmcqueen closes a silent gap in large trademark portfolio searches

The fork now preserves complete owner portfolios instead of quietly losing records when a search returns a lot of marks.

searchintegration

Large trademark portfolios can be central to clearance, diligence and competitor research, but the connector's earlier response could exceed its size limit before all results reached Mike.

This fix retrieves portfolios in smaller batches and sends back a leaner record for each mark: registration details, owner, status, classes, key dates and a short description of the goods or services. It also removes duplicated material that was consuming response space. The change was tested against portfolios of up to 108 records.

The practical result is simple: a large owner search should return the full portfolio rather than a plausible-looking partial answer.

So what Teams using trademark portfolio research should care because incomplete results can distort an assessment of a brand owner's position.

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

1 commit from duncanmcqueen/mike, oldest first. Source extracted verbatim from the harvested git log.

SHA	Subject	Author	Date
`ae52cf95`	fix(mcp): keep trademark owner portfolios within response limits	dwmcqueen	2026-08-13	↗ GitHub
commit body - Page owner search in 10-record lossless pages paced 1s apart and project results into a compact portfolio shape so large portfolios are not truncated by the model-context size limit - Drop the duplicated JSON payload from MCP tool content (it is already in structuredContent) and compact stringifyMcpResult output

SHA

Subject

Author

Date

ae52cf95

fix(mcp): keep trademark owner portfolios within response limits

dwmcqueen

2026-08-13

↗ GitHub

commit body

- Page owner search in 10-record lossless pages paced 1s apart and
  project results into a compact portfolio shape so large portfolios
  are not truncated by the model-context size limit
- Drop the duplicated JSON payload from MCP tool content (it is already
  in structuredContent) and compact stringifyMcpResult output

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-973.md from inside the repo you want the changes in.