No user-submitted related servers found.
Scored across 106 tools
Multiple tools are nearly indistinguishable: get_v1_fbsearch_topsearch, get_v2_fbsearch_topsearch, and get_gql_topsearch all 'search top content by keyword', and media objects are exposed via by_code/by_id/by_url in v1/v2 with and without 'info'. Descriptions rarely explain which version variant to prefer, so an agent would frequently select the wrong endpoint.
All names are lowercase snake_case and start with get_, and most follow get_<version>_<resource>_by_<key>. However, version prefixes are inconsistent (v1/v2/v3/gql/g2/a2) and suffixes like info, chunk, stream, and userstream are not applied uniformly, making the pattern readable but not predictable.
106 tools is an extreme count for an MCP server, especially since many are version duplicates of the same underlying endpoint family. This vastly exceeds the 50+ threshold and will overwhelm agents during tool selection.
For a read-only Instagram data retrieval API, the surface is broad: users, media, stories, highlights, hashtags, locations, music, comments, likers, follower/following lists, and search are all covered. The main gap is the lack of direct media file downloads outside stories and the absence of any write/action endpoints, but those may be out of scope.