Context7
Up-to-date docs for AI agents
제작자가 쓴 설명(영어 원문)
Read-only European city sights: search, nearby, walking routes and itineraries in 6 languages.
이 세 가지는 절대 하나의 점수로 합치지 않아요. 의견, 측정, 인기는 서로 다른 것이니까요.
이 서버는 아직 쉬운 말로 된 요약을 쓰지 못했어요. 아래는 레지스트리에서 읽어 온 사실이에요.
쓰고 있는 AI 앱을 고른 다음 설정을 복사하세요.
Authorization비밀로 표시된 값은 내 계정의 열쇠예요. 절대 다른 사람과 공유하거나 채팅에 붙여 넣지 마세요.
필요한 것
어떤 앱을 쓰세요?
위에서 앱을 고르세요. 기억해 두었다가 다음부터는 버튼에 “…에 설치”라고 표시해요.
이 주소는 요청 header에 키가 필요한데, 커넥터 추가 메뉴에서는 넣을 수 없어요. 대신 Claude Code, Cursor, VS Code를 쓰세요.
터미널에서 이 명령어를 실행하세요
claude mcp add --transport http catalog https://gingerguide.app/mcp --header 'Authorization: your-value'실행하기 전에 빈 값을 채워 주세요.
프로젝트의 .cursor/mcp.json 또는 ~/.cursor/mcp.json에 붙여 넣으세요
{
"mcpServers": {
"catalog": {
"url": "https://gingerguide.app/mcp",
"headers": {
"Authorization": "your-value"
}
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
프로젝트의 .vscode/mcp.json에 붙여 넣으세요
{
"servers": {
"catalog": {
"type": "http",
"url": "https://gingerguide.app/mcp",
"headers": {
"Authorization": "your-value"
}
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
이 주소는 요청 header에 키가 필요한데, 커넥터 추가 메뉴에서는 넣을 수 없어요. 대신 Claude Code, Cursor, VS Code를 쓰세요.
선택한 앱은 이 기기의 쿠키에만 저장돼요. 계정은 필요 없어요.
Authorization 비밀 HTTP header이 설정은 제작자의 레지스트리 정보로 만들었어요. README와 다르면 README를 따르세요.
아직 이 서버의 도구 목록을 읽지 않았어요. 엔드포인트는 차례대로 돌아가며 점검해요.
누구나 볼 수 있는 정보를 점검했고, 코드는 읽지 않았어요. 해당하지 않는 항목은 개수에서 뺐어요. · 2026년 11월 6일 점검
레지스트리 이름이 도메인 gingerguide.app 기준으로 확인되었어요(레지스트리가 도메인 소유를 확인해요)
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
아직 점검하지 않았어요. 엔드포인트는 차례대로 돌아가며 점검해요
아직 점검하지 않았어요. 엔드포인트는 차례대로 돌아가며 점검해요
아직 점검하지 않았어요. 엔드포인트를 점검할 때 도구 목록을 읽어요
레지스트리에서 이 카테고리의 활성 서버 전체와 비교했어요. 스타는 게시자가 저장소를 소유한 경우에만 세요.
해당하는 기본 점검 1개 중 1개를 통과했어요. 코드는 읽지 않았어요. 키나 토큰을 넣기 전에 서버가 어떤 권한을 요구하는지 확인하고, 꼭 필요한 권한만 있는 키를 쓰세요.
입력해야 할 값: Authorization
제작자의 클라우드에서 실행돼요. 설치 없이 인터넷으로 연결하지만, 보내는 내용은 제작자의 서버를 거쳐요.
GitHub 저장소는 마지막으로 – 변경됐고, 최신 버전(2.0.0)은 25일 전 레지스트리에 올라왔어요.
Claude Desktop, Claude Code, Cursor, VS Code처럼 MCP를 지원하는 앱에서 쓸 수 있어요. URL로 연결되기 때문에, 원격 MCP 서버를 지원하는 앱(예: 개발자 모드의 ChatGPT)에서도 쓸 수 있어요.
같은 카테고리에는: Context7, Codebase Memory, Exa
같은 카테고리에서 다른 제작자가 만든 서버예요. 스타 순으로 정렬했어요.
카드 보는 법:★ GitHub 스타i세 기둥: 리뷰 · 점검 · 사용 정도i업데이트 = 최신 버전을 레지스트리에 공개한 날i
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
제작자의 저장소에서 쓰인 그대로 가져왔어요. PopNx MCP가 검토하거나 보증한 내용이 아니에요. HTML, 이미지, 스크립트는 제거했어요. 링크를 누르면 이 사이트를 벗어나요.
[Gingerguide](https://gingerguide.app) is a self-guided audio walking tour app for iPhone and Android. It covers 275 European cities and 4,500+ narrated sights in 46 countries. Each sight has a researched 2 to 3 minute story, read by a lifelike AI voice, in English, Spanish, French, German, Turkish and Chinese. Cities work fully offline.
This repository lets AI assistants and agents use the Gingerguide catalog. It contains:
https://gingerguide.app/mcp. Its 14 tools find sights by interest, historical period or location, recommend sights and cities, compare cities, find day trips, give practical visit facts, and plan walking tours and multi-day itineraries, in six languages.SKILL.md format) for audio tour lookup and one-day city walks.The server is read-only. It needs no API key and no sign-in. An optional free API key raises the daily limit (see Optional free API key).
| MCP endpoint | https://gingerguide.app/mcp |
| Transport | Streamable HTTP (stateless, JSON responses) |
| Auth | None required (optional free API key) |
| MCP Registry name | app.gingerguide/catalog |
| Developer docs | https://gingerguide.app/developers/ |
| OpenAPI (REST) | https://gingerguide.app/api/openapi.json |
| llms.txt | https://gingerguide.app/llms.txt |
The server works without a key. Without a key, it allows 1,000 tool calls per IP address per UTC day. Hosted assistants such as Claude and ChatGPT call from a few shared IP addresses, so their requests get 50,000 per IP address per day. A free key raises the limit to 20,000 tool calls per key per day and shows your usage. It is useful when your agent runs on a shared cloud (Vercel, AWS, Cloudflare). Only tools/call counts. There is no sign-up and no email.
# Create a key (it is shown once; the server stores only its hash)
curl -sS -X POST https://gingerguide.app/api/keys \ -H 'content-type: application/json' -d '{"label":"my-agent"}'
curl -sS https://gingerguide.app/mcp \ -H 'authorization: Bearer ggk_...' \ -H 'content-type: application/json' -H 'accept: application/json' \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
curl -sS https://gingerguide.app/api/keys/me -H 'authorization: Bearer ggk_...' curl -sS -X DELETE https://gingerguide.app/api/keys/me -H 'authorization: Bearer ggk_...'
You can also send the key as X-API-Key. An unknown or revoked key gets HTTP 401. Remove the header to continue without a key. Tool-call responses carry RateLimit and X-RateLimit-Remaining headers. Details: https://gingerguide.app/developers/#api-key.
Every tool is read-only (readOnlyHint: true) and accepts language: en, es, fr, de, tr or zh.
| Tool | What it does | Main arguments |
|---|---|---|
search_attractions | Searches sights in all cities or in one city, by free text, category, interest, historical period, indoor or outdoor, visit length, free entry, or distance from a point. Example: five Byzantine sights in Istanbul. | query, city_id, country, categories, interests, periods, setting, free_only, near_lat, near_lng, radius_km, sort, limit |
find_nearby_attractions | Returns the sights closest to a latitude and longitude, with walking minutes. If no sight is in the radius, it returns the nearest covered city. | lat, lng, radius_km, interests, categories, pace |
get_attraction | Returns one sight: long description, a notable fact, photo, coordinates, map link, visit time and the nearest other sights. In researched cities it adds entry, booking, weekly closing days, dress code and the official site. | attraction_id (for example rome-colosseum) |
get_visit_info | Practical facts for one city: best months and crowd seasons, how to get from the airport, main station or cruise port to the centre, city passes, free sights, what to book ahead, weekly closing days and dress codes. Every fact has a source and a check date. Researched cities only (Rome, Barcelona and Vienna so far); for other cities it says so. | city_id, topics |
search_cities | Finds cities by name (in any of the six languages), country or region. Returns the city id that the other tools need. | query, country, limit |
get_city_guide | Returns one city: overview, headline sights, lesser-known sights, and every sight with category, coordinates and visit time. detail: "full" adds every long description. | city_id, detail |
list_countries | Lists the covered countries with their city ids. | none |
get_catalog_overview | Returns coverage totals, languages, and the exact values that the categories, interests and periods filters accept. | none |
| Tool | What it does | Main arguments |
|---|---|---|
recommend_attractions | Picks sights in one city for the traveler's interests, favourite periods, headline sights or hidden gems, indoor or outdoor, and time available. Each pick says why it was chosen. | city_id, interests, periods, style, setting, available_minutes, exclude_ids |
recommend_destinations | Ranks cities for interests or historical periods, for example "best cities for Byzantine history", and says why. | interests, periods, country, trip_days |
compare_cities | Compares two to four cities side by side: sights, days needed, strongest interests and periods, share of indoor sights, headline sights and the distance between them. Says which city is stronger for each interest or period. | city_ids, interests, periods, trip_days |
find_day_trips | Lists covered cities that make a day trip from a base city, ranked by distance, interests and fame, with what to see there in one day. Sea crossings that need a flight are left out; short ferry trips are marked. | city_id, max_distance_km, interests, periods, hours_on_site |
| Tool | What it does | Main arguments |
|---|---|---|
plan_walking_tour | Builds an ordered walking route in one city for a time budget. Returns each stop with walking minutes, time on site, clock times, and a Google Maps walking link. | city_id, duration_minutes, interests, periods, style, pace, start_lat, start_lng or start_attraction_id, start_time, must_include, weekday |
plan_trip_itinerary | Plans 1 to 7 days in one city. Each day covers one or two walkable areas, with a public transport hop between them. | city_id, days, hours_per_day, interests, periods, style, pace, start_time, start_weekday |
Filter values. Interests: history, architecture, art, museums, religious_sites, archaeology, nature, views, food, local_life, shopping, relaxing, royalty, military, music, literature, science, performing_arts, memorials. Periods: prehistoric, ancient_greek, etruscan, roman, byzantine, medieval, gothic, moorish_islamic, ottoman, renaissance, baroque, georgian_neoclassical, nineteenth_century, art_nouveau, twentieth_century_modern, world_war_ii, cold_war. Style: must_see, balanced, hidden_gems. Pace: relaxed, normal, brisk.
Limits of the data. Interests, periods and indoor or outdoor are derived from each sight's category and description. Walking times are estimates (straight-line distance x 1.3 at the chosen pace). Day-trip and comparison distances are straight-line and come without travel times. With weekday, the planners leave out sights that are closed that day or temporarily closed, in researched cities. Opening hours and ticket prices are not included. Bad input returns an error that names the allowed values.
claude mcp add --transport http gingerguide https://gingerguide.app/mcp
Open Settings → Connectors → Add custom connector. Set the name to Gingerguide and the URL to https://gingerguide.app/mcp.
For older Claude Desktop builds that only read claude_desktop_config.json, use the bridge config in examples/claude_desktop_config.json.
Add this to ~/.cursor/mcp.json (or .cursor/mcp.json in a project). File: examples/cursor-mcp.json.
{
"mcpServers": { "gingerguide": { "url": "https://gingerguide.app/mcp" } } }
Add this to .vscode/mcp.json. File: examples/vscode-mcp.json.
{
"servers": { "gingerguide": { "type": "http", "url": "https://gingerguide.app/mcp" } } }
In a workspace or plan that allows custom connectors (developer mode), add a new connector with the URL https://gingerguide.app/mcp and no authentication.
Any MCP client that supports remote Streamable HTTP servers can connect to https://gingerguide.app/mcp. See examples/curl.md for raw JSON-RPC calls.
The skills/ folder has two skills in the SKILL.md format. Claude, Claude Code and other agents that read this format can load them.
| Skill | Use it when a person... |
|---|---|
find-audio-walking-tour | asks for an audio tour, walking tour or self-guided guide for a city |
plan-city-day-with-gingerguide | asks for a one-day itinerary or walking route in a covered city |
The same skills are also served from the website at https://gingerguide.app/.well-known/agent-skills/index.json.
| Endpoint | Returns |
|---|---|
GET https://gingerguide.app/api/cities.json | All cities with id, country, sight count and guide URL |
GET https://gingerguide.app/api/cities/{id}.json | One city with all its sights |
GET https://gingerguide.app/api/openapi.json | The OpenAPI description |
The cli/ folder has the source of the official command-line client. It is on npm as gingerguide.
npx gingerguide cities
npx gingerguide city rome npx gingerguide countries npx gingerguide health
The files in this repository (docs, configs, skills) are under the MIT License. The catalog data that the server and the API return belongs to Gingerguide. See the [ter
앞부분만 보여 드려요. 나머지는 저장소에서 읽어 보세요