# feat\(mcp\): expose the existing thread search query · PR \#10561 · pingdotgg/t3code

[View on GitCafe](https://git.cafe/pingdotgg/t3code/pull/10561)

Repository: [pingdotgg/t3code](https://git.cafe/pingdotgg/t3code)

Visibility: public

State: merged

Source repository: pingdotgg/t3code

Source branch: agents/mcp-thin/thread-search

Target branch: agents/mcp-thin/preferences

Head commit: 6342c1cb4b4f4484047fb16175163020b27dbab2

[Discussion and reviews](https://git.cafe/pingdotgg/t3code/pull/10561)

[Commits](https://git.cafe/pingdotgg/t3code/pull/10561/commits?format=markdown)

[Files changed](https://git.cafe/pingdotgg/t3code/pull/10561/diffs?format=markdown)

## Description

---

Part 11/16 of the [shared-core and MCP stack](https://github.com/pingdotgg/t3code/stacks/10582). Based on [#10560](https://github.com/pingdotgg/t3code/pull/10560). Next: [#10562](https://github.com/pingdotgg/t3code/pull/10562).

Expose the existing bounded searchThreads query and return only matches from the calling project.

This filters the global top results, so a project can underfill the requested limit. It is not exhaustive or paginated. Project-first SQL pagination and search-service correctness changes belong in a separate service PR.

MCP-only rebuild of [#8721](https://github.com/pingdotgg/t3code/pull/8721), [#8722](https://github.com/pingdotgg/t3code/pull/8722), preserving attribution to Julius Marminge's original work. Original branches remain available for separate service follow-ups.

Validation: The composed stack passes 94 tests across 12 focused files, including shared core MCP and real MCP/V2 integration, attachment intake, project RPC/service contracts, and client model command selection. Server/contracts/shared/client-runtime typechecks and targeted format/lint/diff checks pass. New behavior coverage lives with the shared operation; no per-tool mock suite was added. Current-head CI is shown below.

Layer size: 4 files, +32/-1. No domain-service production implementation or documentation files change in this MCP layer.

Prepared with Codex in the OpenAI agent runtime.