HK Nearby Facilities

Data & APIs

Find nearby Hong Kong public facilities by place keyword or shared live location, starting with public toilets (公廁 / washroom / restroom / toilet) using official government data. Use when the user asks in Cantonese, Chinese, or English things like「大埔墟附近有冇公廁」「沙田站最近廁所喺邊」「旺角 public toilet near me」「附近有冇洗手間」or shares coordinates / live location and wants the nearest facility. This skill currently supports public toilets first and can be extended later with other facilities such as petrol stations.

Install

openclaw skills install hk-nearby-facilities

HK Nearby Facilities

Use the bundled script to search Hong Kong public-facility data by place keyword.

Current scope

  • Public toilets (toilet) via FEHD official JSON endpoints.
  • Future-friendly structure for other facility types later.

Quick start

python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet <place-keywords>

Examples:

python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet 大埔墟
python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet 火炭站
python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet Central --limit 5
python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet 旺角 --accessible-only
python3 /home/jim/.openclaw/workspace/skills/hk-nearby-facilities/scripts/hk_nearby_facilities.py toilet --lat 22.448628 --lng 114.164727 --limit 3

Workflow

  1. Identify the facility type from the request.
    • For now, only toilet is implemented.
  2. Extract the place keywords.
    • Accept district names, station names, estates, landmarks, street names, or short area phrases.
  3. If the user shared live location or latitude+longitude, prefer using the coordinates to rank facilities by physical distance.
  4. Run the script.
  5. Prefer the closest or best-matching few entries rather than a huge list.
  6. Include Google Maps links when they help the user navigate.
  7. If the user asks for accessible / universal toilets, use the relevant flags.
  8. If there is no exact place match, say it is a nearby match instead of pretending it is exact.

Reporting guidance

  • Keep it practical: name, address, district, and map link.
  • Say when the result is approximate or nearby rather than exact.
  • If the query is broad, return the top few likely matches.
  • Mention accessible / universal toilet status when available and relevant.

Data source

Public toilets currently use FEHD official endpoints, including:

  • getMapData.php?type=toilet
  • getAccessibleToilets.php?type=toilet
  • getUniversalToilets.php?type=toilet

Limitations

  • This version is keyword-based, not GPS turn-by-turn routing.
  • It currently supports public toilets only.
  • Oil stations / petrol stations are not included yet because a comparable official government location dataset has not been confirmed.

scripts/

  • scripts/hk_nearby_facilities.py — search nearby Hong Kong facilities by place keyword.