HikerAPI — Instagram MCP
hikerapi-mcp
HikerAPI — 인스타그램 데이터 API를 위한 MCP 서버입니다. npm에서 이용 가능합니다: hikerapi-mcp.
시작 시 HikerAPI OpenAPI 사양에서 MCP 도구를 자동으로 생성하므로, 더 이상 사용되지 않는(deprecated) 엔드포인트를 제외한 모든 GET 엔드포인트가 별도의 수동 래퍼 없이 노출됩니다. HikerAPI는 읽기(GET) 엔드포인트만 제공하며, 서버는 각 엔드포인트를 MCP 도구에 1:1로 매핑합니다 (GET /v2/user/by/username → get_v2_user_by_username).
100개의 무료 API 요청 받기
**이 링크로 가입**하시면 100개의 무료 HikerAPI 요청을 받으실 수 있습니다 — 신용카드가 필요 없습니다. MCP 서버를 연결하고, Claude/Cursor/Codex에서 몇 가지 프롬프트를 테스트하며, 데이터를 결정하기 전에 데이터 품질을 평가하기에 충분한 양입니다.
Related MCP server: DataLikers — Instagram & TikTok MCP
빠른 시작
hikerapi.com/tokens에서 API 키를 받으세요.
AI 어시스턴트에 서버를 추가하세요.
어시스턴트에게 다음과 같이 질문해 보세요:
"@nasa의 인스타그램 프로필을 가져와 줘."
"해시태그
#photography의 최근 게시물 상위 5개를 찾아 줘.""id가 25025320인 사용자의 스토리를 보여 줘."
Claude Code
claude mcp add hikerapi -e HIKERAPI_KEY=your-api-key -- npx -y hikerapi-mcpClaude Desktop
claude_desktop_config.json에 추가하세요:
{
"mcpServers": {
"hikerapi": {
"command": "npx",
"args": ["-y", "hikerapi-mcp"],
"env": {
"HIKERAPI_KEY": "your-api-key"
}
}
}
}Cursor / Windsurf
Claude Desktop과 동일한 형식입니다 — 앱의 MCP 설정 파일 내 mcpServers 블록 아래에 넣으세요.
Zed
~/.config/zed/settings.json에 추가하세요:
{
"context_servers": {
"hikerapi": {
"command": "npx",
"args": ["-y", "hikerapi-mcp"],
"env": {
"HIKERAPI_KEY": "your-api-key"
}
}
}
}OpenAI Codex
~/.codex/config.toml에 추가하세요:
[mcp_servers.hikerapi]
command = "npx"
args = ["-y", "hikerapi-mcp"]
[mcp_servers.hikerapi.env]
HIKERAPI_KEY = "your-api-key"도구
도구는 라이브 HikerAPI OpenAPI 사양에서 시작 시 생성되므로, 목록은 항상 현재 API와 일치합니다. 이 글을 작성하는 시점 기준으로 약 100개 이상의 도구가 다음 그룹으로 나뉩니다:
그룹 | 도구 수 | 예시 |
사용자 프로필 | 36 |
|
게시물 상세 | 20 |
|
검색 | 13 |
|
해시태그 | 7 |
|
스토리 | 7 |
|
위치 | 7 |
|
오디오, 공유, 하이라이트, 댓글 | ~10 |
|
각 도구 이름은 해당 엔드포인트를 반영합니다 (GET /v2/user/by/username → get_v2_user_by_username). 어시스턴트는 MCP를 통해 tools/list를 호출하여 매개변수 스키마가 포함된 최신 전체 목록을 가져올 수 있습니다. Legacy 및 System 그룹은 기본적으로 제외됩니다.
설정
변수 | 설명 | 필수 여부 |
| HikerAPI 액세스 키 ( | 예 |
| 기본 URL. 기본값: | 아니오 |
| OpenAPI 사양 URL. 기본값: | 아니오 |
| 화이트리스트: 이 태그가 포함된 작업만 포함 (쉼표로 구분) | 아니오 |
| 블랙리스트: 제외할 추가 태그 (기본값 | 아니오 |
| API 호출당 요청 시간 제한. 기본값: | 아니오 |
| 시작 시 사양 가져오기 시간 제한. 기본값: | 아니오 |
| 각 API 응답에서 읽을 최대 바이트 수. 기본값: | 아니오 |
| OpenAPI 사양에서 읽을 최대 바이트 수. 기본값: | 아니오 |
Legacy 및 System 태그는 기본적으로 제외됩니다. 더 이상 사용되지 않는 작업도 건너뜁니다.
HIKERAPI_URL이 api.hikerapi.com 또는 api.instagrapi.com 이외의 호스트를 가리키는 경우, 서버는 시작 시 경고를 출력합니다. 키가 해당 주소로 전송되므로, 직접 호스팅하거나 프록시된 HikerAPI에만 사용하십시오.
예시 — 가장 일반적인 그룹만 노출:
"env": {
"HIKERAPI_KEY": "...",
"HIKERAPI_TAGS": "User Profile,Post Details,Search,Hashtags,Stories"
}작동 방식
AI Assistant ←stdio→ hikerapi-mcp ──https──> api.hikerapi.com
│
└─ fetches /openapi.json once on startup,
builds one MCP tool per GET endpoint도구 인수는 엔드포인트의 query 및 path 매개변수에 매핑됩니다. 응답 본문은 그대로(JSON 텍스트) 반환됩니다. 2xx가 아닌 응답은 HTTP 상태 코드와 본문을 포함한 도구 오류로 표시됩니다.
개발
git clone https://github.com/subzeroid/hikerapi-mcp.git
cd hikerapi-mcp
npm install
npm run build
HIKERAPI_KEY=your-key node dist/index.js감시 모드에서 실행:
HIKERAPI_KEY=your-key npm run dev테스트 실행 (단위 테스트 + 로컬 모의 서버에 대한 stdio 스모크 테스트, 네트워크/API 키 불필요):
npm test라이선스
MIT
Available Tools
44 toolsget_g2_location_by_idARead-only
Get an Instagram location by id: name, latitude/longitude, category and description. Use get_v1_fbsearch_places or get_v1_location_search to find the id first, and the location media tools for posts made there. Live request to Instagram, billed per call. (GET /g2/location/by/id)
Inputs: id (required).
get_g2_user_followersARead-only
List accounts that follow an Instagram user, one page per call. Use it to page through the follower list; use get_v1_user_search_followers to look for specific names among them and get_g2_user_following for accounts the user follows. Pass user_id; for the next page pass the cursor from the previous response as page_id. Live request to Instagram, one billed request per page. (GET /g2/user/followers)
Inputs: user_id (required), page_id.
get_g2_user_followingARead-only
List accounts an Instagram user follows, one page per call. Use it to page through the following list; use get_v1_user_search_following to look for specific names in it and get_g2_user_followers for the user's followers. Pass user_id; for the next page pass the cursor from the previous response as page_id. Live request to Instagram, one billed request per page. (GET /g2/user/following)
Inputs: user_id (required), page_id.
get_gql_comment_likers_chunkARead-only
Get users who liked one comment on an Instagram post, one page per call. Pass comment_id and media_id; paginate with end_cursor from the previous response. Use get_v2_media_likers for likers of the post itself. Live request to Instagram, billed per call. (GET /gql/comment/likers/chunk)
Inputs: media_id, comment_id, end_cursor.
get_gql_media_usertagsARead-only
Get the users tagged in Instagram videos; pass up to 10 media ids in media_ids. Use it for reels and videos, because the get_v2_media_info_* tools do not return user tags for video. Live request to Instagram, billed per call. (GET /gql/media/usertags)
Inputs: media_ids.
get_gql_user_aboutARead-only
Get the "About this account" panel for an Instagram user id: verified status, country of registration, account creation date and former usernames. Use it for account provenance checks; use get_v2_user_by_id for the regular profile. Live request to Instagram, billed per call. (GET /gql/user/about)
Inputs: id (required).
get_gql_user_mediasARead-only
List an Instagram user's posts, one page per call. Use it to read an account's feed; use get_v2_user_clips for reels only and get_v2_user_tag_medias for posts the user is tagged in. Pass user_id; for the next page pass the cursor from the previous response as profile_grid_items_cursor. flat=true flattens nested media objects into a single list. Live request to Instagram, billed per call. (GET /gql/user/medias)
Inputs: user_id (required), flat, profile_grid_items_cursor.
get_gql_user_repostsARead-only
List content an Instagram user has reposted, one page per call. Use get_gql_user_medias for the user's own posts. Pass user_id; for the next page pass the cursor from the previous response as repost_next_max_id. flat=true returns a simple items list. Live request to Instagram, billed per call. (GET /gql/user/reposts)
Inputs: user_id (required), flat, repost_next_max_id.
get_v1_fbsearch_placesARead-only
Search Instagram places by name (query), optionally with lat and lng. Use it to find a location id for get_g2_location_by_id and the location media tools; use get_v1_location_search when you only have coordinates. Live request to Instagram, billed per call. (GET /v1/fbsearch/places)
Inputs: query (required), lat, lng.
get_v1_hashtag_by_nameARead-only
Get an Instagram hashtag by exact name (no # and no special characters). Use it for the hashtag's own info; use get_v2_hashtag_medias_top or get_v2_hashtag_medias_recent for its posts and get_v1_search_hashtags when unsure of the exact tag. Live request to Instagram, billed per call. (GET /v1/hashtag/by/name)
Inputs: name (required).
get_v1_hashtag_medias_clips_chunkARead-only
List reels under an Instagram hashtag, one page per call. Use get_v2_hashtag_medias_top or get_v2_hashtag_medias_recent for all post types and get_v2_fbsearch_reels to search reels by free text. Pass name without #; paginate with max_id from the previous response. Live request to Instagram, billed per call. (GET /v1/hashtag/medias/clips/chunk)
Inputs: name (required), max_id.
get_v1_highlight_by_urlARead-only
Get one Instagram story highlight by its link. For share links that contain /s/, call get_v1_share_by_url first to resolve them. Use get_v2_highlight_by_id when you have the id. Live request to Instagram, billed per call. (GET /v1/highlight/by/url)
Inputs: url (required).
get_v1_location_medias_recent_chunkARead-only
List recent posts made at an Instagram location, one page per call. Pass location_pk; paginate with max_id from the previous response. Use get_v1_location_medias_top_chunk for top posts. Live request to Instagram, billed per call. (GET /v1/location/medias/recent/chunk)
Inputs: location_pk (required), max_id.
get_v1_location_medias_top_chunkARead-only
List top posts made at an Instagram location, one page per call. Pass location_pk; paginate with max_id from the previous response. Use get_v1_location_medias_recent_chunk for the newest posts. Live request to Instagram, billed per call. (GET /v1/location/medias/top/chunk)
Inputs: location_pk (required), max_id.
get_v1_location_searchARead-only
Find Instagram locations at a coordinate: pass lat and lng. Use get_v1_fbsearch_places to search places by name instead. Live request to Instagram, billed per call. (GET /v1/location/search)
Inputs: lat (required), lng (required).
get_v1_search_hashtagsARead-only
Search Instagram hashtags by keyword. Use it when unsure of the exact tag; use get_v1_hashtag_by_name for one known hashtag. Live request to Instagram, billed per call. (GET /v1/search/hashtags)
Inputs: query (required).
get_v1_search_musicARead-only
Search Instagram music tracks by keyword (song or artist). Use get_v2_track_by_id for one known track. Live request to Instagram, billed per call. (GET /v1/search/music)
Inputs: query (required).
get_v1_user_search_followersARead-only
Search within an Instagram user's followers by name or username. Use it to check whether specific accounts follow a user; use get_g2_user_followers to list all followers. Pass user_id and query; force=true skips the account privacy check. Live request to Instagram, billed per call. (GET /v1/user/search/followers)
Inputs: query (required), user_id (required), force.
get_v1_user_search_followingARead-only
Search within the accounts an Instagram user follows by name or username. Use it to check whether a user follows specific accounts; use get_g2_user_following to list all of them. Pass user_id and query; force=true skips the account privacy check. Live request to Instagram, billed per call. (GET /v1/user/search/following)
Inputs: query (required), user_id (required), force.
get_v2_fbsearch_accountsARead-only
Search Instagram accounts by keyword. Use it to find a profile when you do not know the exact handle; use get_v2_user_by_username for an exact handle. Paginate with page_token from the previous response. Live request to Instagram, billed per call. (GET /v2/fbsearch/accounts)
Inputs: query (required), safe_int, page_token.
get_v2_fbsearch_reelsARead-only
Search Instagram reels by keyword. Use it to find short videos about a topic; use get_v1_hashtag_medias_clips_chunk for reels under one specific hashtag. Paginate with reels_max_id and rank_token from the previous response. Live request to Instagram, billed per call. (GET /v2/fbsearch/reels)
Inputs: query (required), safe_int, rank_token, reels_max_id.
get_v2_fbsearch_topsearchARead-only
Search Instagram for top content by keyword. Use it for a broad first look; use get_v2_fbsearch_accounts to search only accounts, get_v2_fbsearch_reels only reels and get_v1_search_hashtags only hashtags. Paginate with next_max_id from the previous response. Live request to Instagram, billed per call. (GET /v2/fbsearch/topsearch)
Inputs: query (required), safe_int, next_max_id.
get_v2_hashtag_medias_recentARead-only
List recent posts under an Instagram hashtag, one page per call. Use get_v2_hashtag_medias_top for top posts and get_v1_hashtag_medias_clips_chunk for reels only. Pass name without #; for the next page pass next_page_id from the previous response as page_id. Live request to Instagram, billed per call. (GET /v2/hashtag/medias/recent)
Inputs: name (required), page_id, safe_int.
get_v2_hashtag_medias_topARead-only
List top posts under an Instagram hashtag, one page per call. Use get_v2_hashtag_medias_recent for the newest posts and get_v1_hashtag_medias_clips_chunk for reels only. Pass name without #; for the next page pass next_page_id from the previous response as page_id. Live request to Instagram, billed per call. (GET /v2/hashtag/medias/top)
Inputs: name (required), page_id, safe_int.
get_v2_highlight_by_idARead-only
Get one Instagram story highlight by id. Use get_v2_user_highlights to list a user's highlights and get_v1_highlight_by_url when you have a link. Live request to Instagram, billed per call. (GET /v2/highlight/by/id)
Inputs: id (required), safe_int.
get_v2_media_commentsARead-only
Get comments on an Instagram post, about 15 per call. Pass the media id; for the next page pass the page id from the previous response as page_id. Use get_v2_media_comments_replies for replies under one comment and get_v2_media_likers for users who liked the post. Live request to Instagram, one billed request per page. (GET /v2/media/comments)
Inputs: id (required), page_id, safe_int, can_support_threading.
get_v2_media_comments_repliesARead-only
Get replies under one comment of an Instagram post. Use it to read a thread; use get_v2_media_comments for the top-level comments. Pass media_id and comment_id (both from the comments tool); paginate with min_id from the previous response. Live request to Instagram, billed per call. (GET /v2/media/comments/replies)
Inputs: media_id (required), comment_id (required), min_id, safe_int.
get_v2_media_info_by_codeARead-only
Get an Instagram post or reel by its shortcode (the part after /p/ or /reel/ in the link). Use get_v2_media_info_by_url when you have a full link and get_v2_media_info_by_id for a numeric media id. Returns the media object, or 404 for deleted or unavailable posts; promoted (ad) posts may also return 404. User tags are not included for videos, use get_gql_media_usertags for those. Live request to Instagram, billed per call. (GET /v2/media/info/by/code)
Inputs: code (required), safe_int.
get_v2_media_info_by_idARead-only
Get an Instagram post or reel by numeric media id. Use get_v2_media_info_by_code for a shortcode and get_v2_media_info_by_url for a full link. Returns the media object, or 404 for deleted or unavailable posts; promoted (ad) posts may also return 404. User tags are not included for videos, use get_gql_media_usertags for those. Live request to Instagram, billed per call. (GET /v2/media/info/by/id)
Inputs: id (required), safe_int.
get_v2_media_info_by_urlARead-only
Get an Instagram post or reel by its link (instagram.com/p/... or /reel/...). Use get_v2_media_info_by_code for a bare shortcode. Returns the media object, 404 for deleted or unavailable posts and 400 for links that are not posts: use get_v2_story_by_url for story links and get_v1_share_by_url for /s/ share links. Promoted (ad) posts may also return 404. Live request to Instagram, billed per call. (GET /v2/media/info/by/url)
Inputs: url (required), safe_int.
get_v2_media_likersARead-only
Get users who liked an Instagram post, by media id. Use get_gql_comment_likers_chunk for likers of a single comment and get_v2_media_comments for commenters. Live request to Instagram, billed per call. (GET /v2/media/likers)
Inputs: id (required), safe_int.
get_v2_story_by_idARead-only
Get one Instagram story by its id. Use get_v2_story_by_url when you have a story link and get_v2_user_stories to list a user's active stories. Live request to Instagram, billed per call. (GET /v2/story/by/id)
Inputs: id (required), safe_int.
get_v2_story_by_urlARead-only
Get one Instagram story by its link. For share links that contain /s/, call get_v1_share_by_url first to resolve them. Use get_v2_story_by_id when you have the id. Live request to Instagram, billed per call. (GET /v2/story/by/url)
Inputs: url (required), safe_int.
get_v2_track_by_idARead-only
Get an Instagram music track by track_id. Use get_v1_search_music to find tracks by keyword. For the next page pass next_page_id from the previous response as page_id. Live request to Instagram, billed per call. (GET /v2/track/by/id)
Inputs: track_id (required), page_id, safe_int.
get_v2_user_by_idARead-only
Get an Instagram profile by numeric user id. Use it when you already have the id from another tool; use get_v2_user_by_username when you only have a handle. Returns the same user object. Live request to Instagram, billed per call. (GET /v2/user/by/id)
Inputs: id (required), safe_int.
get_v2_user_by_usernameARead-only
Get an Instagram profile by username (without @). Use it when you have a handle; use get_v2_user_by_id when you already know the numeric user id (faster), and get_gql_user_about for the "About this account" panel. Returns the user object; pass its pk as the user id to the other user tools. Live request to Instagram, one billed request per call. (GET /v2/user/by/username)
Inputs: username (required), safe_int.
get_v2_user_clipsARead-only
List an Instagram user's reels (short videos), one page per call. Use it for reels only; use get_gql_user_medias for all posts. Pass user_id; for the next page pass next_page_id from the previous response as page_id. Live request to Instagram, billed per call. (GET /v2/user/clips)
Inputs: page_id, user_id, safe_int.
get_v2_user_highlightsARead-only
List an Instagram user's story highlights (saved story collections) by user id, one page per call. Use get_v2_highlight_by_id to open one highlight and get_v2_user_highlights_by_username when you only have a handle. For the next page pass next_page_id from the previous response as page_id (amount is ignored). Live request to Instagram, billed as 2 requests per call. (GET /v2/user/highlights)
Inputs: user_id (required), force, amount, page_id, +1 more.
get_v2_user_highlights_by_usernameARead-only
List an Instagram user's story highlights by username, one page per call. Use it only when you have a handle and not the user id: it is slower than get_v2_user_highlights and billed as 3 requests per call instead of 2. For the next page pass next_page_id from the previous response as page_id. Live request to Instagram. (GET /v2/user/highlights/by/username)
Inputs: username (required), force, amount, page_id, +1 more.
get_v2_user_storiesARead-only
Get an Instagram user's active stories by user id. Use it when you have the id; use get_v2_user_stories_by_username when you only have a handle and get_v2_user_highlights for saved highlight collections. force=true skips the account privacy check. Live request to Instagram, billed as 2 requests per call. (GET /v2/user/stories)
Inputs: user_id (required), force, safe_int.
get_v2_user_stories_by_usernameARead-only
Get an Instagram user's active stories by username. Use it only when you have a handle and not the user id: it is slower than get_v2_user_stories and billed as 3 requests per call instead of 2. Live request to Instagram. (GET /v2/user/stories/by/username)
Inputs: username (required), force, safe_int.
get_v2_user_suggested_profilesARead-only
Get the accounts Instagram suggests as related to a given user. Use it to find similar accounts; use get_v2_fbsearch_accounts to search accounts by keyword instead. Pass user_id; expand_suggestion=true returns more detail per account. Live request to Instagram, billed per call. (GET /v2/user/suggested/profiles)
Inputs: user_id (required), safe_int, expand_suggestion.
get_v2_user_tag_mediasARead-only
List posts in which an Instagram user is tagged, one page per call. Use it to see who features an account; use get_gql_user_medias for the user's own posts. Pass user_id; for the next page pass next_page_id from the previous response as page_id. Live request to Instagram, billed per call. (GET /v2/user/tag/medias)
Inputs: page_id, user_id, safe_int.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 44 tools
The tools cover many distinct Instagram resources and actions, and descriptions explicitly cross-reference alternatives such as by id vs username vs URL vs shortcode, or top vs recent vs clips. While no two tools are truly interchangeable, the sheer number of near-variants increases cognitive load and misselection risk.
All tools use a get_ + lowercase snake_case convention, making the pattern predictable. Minor inconsistency comes from mixed API-version prefixes (v1, v2, gql, g2) and compound tokens like fbsearch, usertags, and clips_chunk, but the naming remains readable and largely consistent.
44 tools far exceeds the typical 3-15 range and includes many access-method variants that could be consolidated into fewer flexible tools. While each maps to a real endpoint, the set is heavy and likely to overwhelm tool selection.
The surface covers users, media, comments, likes, stories, highlights, search, hashtags, locations, music, followers/following, suggestions, and account provenance, giving broad read-only Instagram coverage. Gaps exist for less common data such as live videos, saved media, or media by music track, but core workflows are covered.
Maintenance
Related MCP Connectors
Hosted MCP server for DataLikers — Instagram & TikTok data API. 51 tools: Instagram user search by demographics (gender/age/race/country/city), profiles, bulk lookup, engagement, posts & reels, comments, hashtags, locations, stories, highlights, music, business accounts, top users; TikTok users, videos, comments, hashtags, playlists and top charts. Streamable HTTP, Bearer API key. Free tier: 100 requests on signup at https://datalikers.com/p/1by27bwg
Instagram data over MCP: profiles, posts, reels, stories, highlights and search as JSON
Unified social scraper MCP — profiles, posts, videos across ~40 platforms via Streamable HTTP.
Hosted Instagram MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
Related MCP Servers
- AlicenseAqualityBmaintenanceMCP server for LamaTok TikTok data API — auto-generates 20+ tools from the live OpenAPI spec (users, videos, hashtags, music, comments). Local stdio via npx -y lamatok-mcp, requires only LAMATOK_KEY.2344 npm1MIT
- AlicenseNot gradedqualityCmaintenanceMCP server for DataLikers — 51 tools for Instagram & TikTok data: Instagram user search by demographics (gender/age/race/country/city), profiles, engagement, posts, comments, hashtags, locations, stories, business accounts; TikTok users, videos, comments, hashtags, playlists. Local stdio via npx -y datalikers-mcp, requires only DATALIKERS_API_KEY.53 npm9MIT
- AlicenseAqualityCmaintenance16 Instagram creator tools as an MCP server — Reels/Story/carousel downloaders, engagement audit, hashtag search, Reels hook generator, best-time-to-post and content calendar. Wraps instapdown.com public API — no auth required.1651 npm2MIT
- AlicenseNot gradedqualityCmaintenanceMCP server for downloading Instagram content (videos, reels, audio, carousels) using yt-dlp, with tools for downloading media and fetching metadata. Supports stdio and HTTP transports.MIT