Google Search Console(GSC)을 AI 어시스턴트와 연결하여 자연어 대화를 통해 SEO 데이터를 분석할 수 있게 하는 모델 컨텍스트 프로토콜(MCP) 서버입니다. Claude Desktop, Cursor, Codex CLI, Gemini CLI, Antigravity 및 기타 MCP 호환 클라이언트와 호환됩니다.

설정 건너뛰기, 더 보기 보다 고급 호스팅 버전으로, 원클릭 로그인, 추가 GA4 도구. Claude Desktop, Claude Code, Claude.ai, Codex, 커서 및 모든 MCP 클라이언트에서 작동합니다. 좌석 수는 단 100석입니다. → 고급 GSC MCP (호스팅)

---

새로운 소식

[0.4.1] — 2026년 9월

• check_indexing_issues 동시 실행 — #31과 동일한 수정으로, 10-URL 배치 시 sc-domain:* 속성에 대해 타임아웃되지 않습니다. 감사합니다@kuldiph. (#55)
• **get_search_by_page_query에서 남은 죽은 orderBy**를 제거했습니다 (검색 분석 API가 이를 무시함). 감사합니다@kuldiph. (#55)

[0.4.0] — 2026년 9월

• 리치 결과 이슈가 이제 드러난다 — inspect_url_enhanced와 batch_url_inspection는 올바른 API 경로에서 리치 결과 이슈를 읽어, 마크업 문제가 조용히 삭제되는 대신 보고됩니다. 감사합니다@patrickweh. (#46, #48)

Google Search Console MCP Server for SEOs

A Model Context Protocol (MCP) server that connects Google Search Console (GSC) to AI assistants, allowing you to analyze your SEO data through natural language conversations. Works with Claude Desktop, Cursor, Codex CLI, Gemini CLI, Antigravity, and any other MCP-compatible client.

Skip setup, get more. A more advanced hosted version — one-click sign-in, added GA4 tools. Works with Claude Desktop, Claude Code, Claude.ai, Codex, Cursor, and any MCP client. Only 100 seats. → Advanced GSC MCP (hosted)

---

What’s New

[0.4.1] — September 2026

• check_indexing_issues runs concurrently — same fix as #31, so it no longer times out on 10-URL batches against sc-domain:* properties. Thanks @kuldiph. (#55)
• Removed a leftover dead orderBy in get_search_by_page_query (the Search Analytics API ignores it). Thanks @kuldiph. (#55)

[0.4.0] — September 2026

• Rich-result issues now surface — inspect_url_enhanced and batch_url_inspection read rich-result issues from the correct API path, so markup problems are reported instead of silently dropped. Thanks @patrickweh. (#46, #48)
• batch_url_inspection runs concurrently — 10-URL batches no longer time out on sc-domain:* properties. Thanks @remotesensei. (#31)
• compare_search_periods deltas fixed — Period 1 is now the analyzed period and Period 2 the baseline, so growth reads as growth (not decline). Thanks @JKdreaming. (#42)
• get_advanced_search_analytics sorting works — sort_by/sort_direction are now applied instead of silently ignored. Thanks @kuldiph. (#54)
• Hardened error handling — replaced bare except: clauses so fatal signals aren’t swallowed. Thanks @kuldiph. (#53)
• Note: batch_url_inspection renames the per-URL verdict field to index_verdict to distinguish it from the rich-result verdict.

[0.3.3] — July 2026

• Fixed fresh installs broken by mcp 2.0 — pinned mcp[cli]<2.0.0. The mcp SDK 2.0.0 (released 2026-07-28) removed the mcp.server.fastmcp module, so every fresh uvx mcp-search-console install crashed on startup with ModuleNotFoundError: No module named 'mcp.server.fastmcp'. New installs now resolve a working 1.x SDK again — no --with "mcp<2" workaround needed.

[0.3.2] — April 2026

• OAuth browser flow fixed for uvx — removed the isatty block that prevented the browser login window from opening when running as an MCP subprocess on macOS. OAuth now works out of the box with uvx, no manual terminal run needed.