Skip to main content
Glama
lovelyquality

korail-mcp

by lovelyquality
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Hybrid

Related Servers

Alternatives to korail-mcp

No user-submitted related servers found.

    Related Servers

    • korean-public-data-mcp

      SongT-50
      A
      license
      Not graded
      quality
      D
      maintenance
      Enables AI to query real-time Korean public data including weather, real estate prices, air quality, economic indicators, and business registration via natural language.
      1
      MIT

    • Korea Stats MCP

      Dayoooun
      A
      license
      A
      quality
      B
      maintenance
      Enables natural language querying of Korean statistical data from KOSIS, including population, employment, GDP, housing prices, and more, with support for regional and trend analysis.
      8
      8 npm
      16
      MIT

    • Korean Stats MCP

      nagyeop
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI assistants to retrieve real-time official Korean statistics from the KOSIS database through natural language queries, with source citations.
      MIT

    • kosis-mcp

      updown256
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables querying Korean official statistics from KOSIS via natural language in MCP clients like Claude Desktop, wrapping the KOSIS OpenAPI for search, data retrieval, and metadata exploration.
      MIT

    • OpenAPI Korea MCP Server

      ggumgil
      A
      license
      Not graded
      quality
      D
      maintenance
      Enables access to Korean public data services through OpenAPI integration. Supports querying government datasets like parking information in Sejong City through natural language interactions.
      MIT
    Deploy Server
    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

    TDQS

    B3.1/5.0

    Scored across 98 tools

    Disambiguation2/5

    There are several clusters of tools with unclear boundaries, such as get_station_facilities, get_accessible_facilities, get_station_facilities_detail, and the get_urban_* facility tools, which all cover similar physical station amenities from different sources. More seriously, get_mainline_carriage overlaps heavily with get_mainline_station_per, and get_wide_area_carriage overlaps with get_wide_rail_station_per, so an agent can easily select the wrong endpoint.

    Naming Consistency3/5

    The set mostly uses English snake_case with get_/search_/list_/decode_ prefixes, so the broad pattern is readable. However, analogous tools are named inconsistently: get_mainline_carriage vs get_mainline_station_per, get_transport_stat_codes vs get_train_codes, and search_loading_time_adjustment vs list_standard_loading_time all use different verb/object conventions for similar lookup operations.

    Tool Count1/5

    98 tools is an extreme size for a single MCP server and far exceeds the 50+ threshold for a coherence failure. Even if KORAIL has many datasets, the server should be split into focused sub-servers by domain such as urban rail, mainline operations, freight, station facilities, or corporate administration.

    Completeness2/5

    Read-only coverage is broad, but there are significant gaps for a KORAIL information server: current passenger fare lookup is missing, real-time train status/delay information is absent, and some datasets are only snapshots or single-day samples such as get_train_run_history. The many duplicated station and facilities datasets also make it unclear whether a query is fully answered without checking multiple tools.

    Maintenance

    ActivitySlowing
    ResponsivenessNo issues

    Resources