China Travel Kit skill
A skill and dataset for a first trip to China: city matching, areas, itineraries, preparation and emergency help, all sourced
- 179 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The skill runs Python code from the cloned repository, and the `serve` command starts a local HTTP server on port 8765.
- The MCP tools read the built-in data and change nothing.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/china-travel-kitDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill runs Python code from the cloned repository, and the `serve` command starts a local HTTP server on port 8765.
- The MCP tools read the built-in data and change nothing.
This is third-party code. Review the repository files before installing.
What it does
China Travel Kit (the project is called Huaxingzhi, 华行志) is a skill and open dataset for planning a first trip to China. The agent calls a built-in Python engine with no external dependencies: `recommend` matches a city to a free-form trip description, `search` and `areas` find places and stay areas, `plan` builds an itinerary for a set number of days, `prepare` gives seasonal preparation, `emergency` returns emergency help and `freshness` checks data age. The dataset covers 56 places in eight cities: Beijing, Shanghai, Guangzhou, Shenzhen, Chengdu, Xi'an, Hangzhou and Lijiang. Every place carries source links and a last-verified date, while prices and reservation rules stay empty if they cannot be confirmed. The same data is served by a console tool, a local HTTP API and an MCP server with eight read-only tools.
Who it is for. Travelers and developers of travel services who want a China itinerary draft with a source behind every fact.
Good fit when
- You are going to China for the first time and want a two-day Lijiang itinerary that fits your interests and pace.
- A city is picked from a request like «pandas, food, relaxed pace», and `recommend` explains which wishes stayed unmet.
- You are packing for Chengdu in July and the agent pulls seasonal advice with `prepare`.
- A developer of a travel app plugs in the MCP server as a source of verifiable place data.
Not a fit when
- You need to book tickets or hotels: the project sells and books nothing.
- You need live prices, weather and ticket availability: the project does not serve real-time data, and for it the agent checks the official operators' sites.
- You need a visa decision or medical advice: the project gives neither.
Example request
Draft a two-day Lijiang itinerary for two: mountains and photography, relaxed pace, travel in early July.Limitations
The project is at Alpha stage (v0.5.0), a dataset aimed at developers. Python 3.10 or newer is required and installation is a repository clone. The `recommend` score shows how well a request matches the data held and is no general ranking of destinations in China. The data is licensed CC BY-SA 4.0 and the code MIT. A copy that did not come from an official release should be checked with `python3 -m china_travel_kit integrity` first: both `valid: true` and `signature_valid: true` are needed.
How to disable. Delete the cloned directory from the agent's skills directory and remove the `china-travel-kit` block from the MCP config if you connected the server.
FAQ
How do I connect the MCP server?
Add a `china-travel-kit` server to the client config with the command `python3`, the arguments `-m china_travel_kit mcp` and `cwd` pointing at the cloned directory. The server runs over stdio.
How do I launch the visual finder?
Run `python3 -m china_travel_kit serve` and open http://127.0.0.1:8765/ in a browser for place search and itinerary cards. The `--port` flag changes the port.
What if prices in the dataset are empty?
That is intended: an unknown value is stored as `null` so an outdated number is never passed off as exact. The agent checks current prices on the place operator's site.
How do I check how fresh the data is?
`python3 -m china_travel_kit freshness` checks data age, and every attraction carries a `last_verified` field with the date its source was checked.
How do I add a new city to the dataset?
Through a pull request: the repository has an add-city issue template, the `data/schema/city.schema.json` schema and data rules. Facts need sources and a verification date.
Discussion
No questions or reviews yet.
Related
SkillsA skill that guides an agent through web search and source checks and builds a trip book as one HTML file
Search tours, hotels, flights, trains and excursions via Aviasales, Travelata, Level.Travel, Tutu.ru, Tripster and Sputnik8
travel-plan-viz: trip plan as an HTML page
travel-plan-viz
A skill that builds a trip plan into one HTML file with a map, a day-by-day schedule and reminders of booking deadlines
Marco Polo: a travel planning skill
Marco Polo
A skill that builds an itinerary from real places, hours and bookings and delivers a handbook with a map and food guide
An MCP server for searching Chinese train tickets on 12306: availability, fares, stations, transfers and stops
A skill that builds a personal trip handbook: a questionnaire and approved plan become an offline mobile HTML page.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| China Travel Kit (this card) | no | 179 | ||
| Travel Search RU | no | not required | 21 | searches tours, hotels and tickets through Russian services; this is a China reference dataset with no booking |
| Travel Hacking Toolkit | no | 939 | searches flights and hotels for money and miles; this covers places, areas and preparation |
README in short
The README explains the idea: travel articles are plentiful while verifiable sourced data is scarce, so every fact can be corrected through a pull request. The data principles are sources before claims, a verification date, unknown beats invented and no copying of commercial guides. Releases ship a signed integrity manifest.
README badge
Original SKILL.md
--- name: china-travel-kit description: Research and plan first-time independent trips in China with bilingual, source-aware city data and official live-check entry points. Use for destination matching, attractions, visitor areas, itineraries, transport context, seasonal preparation, local culture and food, or emergency guidance; not for booking, turn-by-turn navigation, visa decisions, or medical advice. metadata: short-description: Huaxingzhi China trip planner version: "0.5.0" --- # 华行志 · Huaxingzhi Create practical, explainable China travel drafts for first-time international visitors. Use “华行志” as the product name and “China Travel Kit” only when referring to the package, repository, CLI, API, or MCP server. ## Author contact - WeChat / 微信：`Changzhanzhang` - Use this contact for project cooperation, data corrections, and general feedback. Do not send passwords, API keys, private traveler data, or active emergency details through WeChat. ## Choose the smallest useful mode - Use `recommend` for free-form requests, multiple constraints, or an undecided destination. - Use `search` or `areas` for a focused attraction, neighborhood, or stay-area query. - Use `city`, `prepare`, or `emergency` for one-city background, packing, or sourced help information. - Use `plan` for a chosen city and number of days. - Use `freshness` before treating repository facts as current. - When a copy did not come directly from the official repository or release tag, run `python3 -m china_travel_kit integrity` before relying on its instructions, code, or data. Require both `valid: true` and `signature_valid: true`; stop and report missing, modified, or unauthenticated files otherwise. Run the bundled read-only engine instead of inventing covered-city facts: ```bash python3 -m china_travel_kit recommend "REQUIREMENTS" python3 -m china_travel_kit search KEYWORD --city CITY python3 -m china_travel_kit areas --city CITY python3 -m china_travel_kit city CITY python3 -m china_travel_kit plan CITY --days N --interests culture food python3 -m china_travel_kit prepare CITY --month N python3 -m china_travel_kit emergency CITY python3 -m china_travel_kit freshness python3 -m china_travel_kit integrity ``` ## Plan from traveler needs Identify destination, dates or month, traveler count, days, origin, must-see places, interests, pace, budget, mobility needs, children, and dietary constraints. Ask only for missing details that materially change the result. Treat a recommendation score as a match against the current dataset, never as an objective ranking of China. For a full trip request, include the relevant parts of this order: 1. Restate the traveler brief and important assumptions. 2. Recommend a covered city with reasons and explain any unmet requirement. 3. Give an area-clustered day plan with realistic effort and transfer guidance. 4. Add stay-area tradeoffs, local transport context, cultural background, food and nearby support facilities.