Local Place Search logo

Local Place Search

Organization
brave
local-place-search

USE FOR finding places in the physical world - businesses, POIs, street addresses, cities and streets. Results carry address, coordinates, rating, opening hours and phone, so basic details need no follow-up call. Standalone - no POI IDs or prior web search required; the IDs it returns work with local-pois and local-descriptions. Locate by coordinates or a location string, or omit both to search globally. Omit the query to browse an area. Max 100 results.

Overview

Publisherbrave
Repositorybrave-search-skills
Skill namelocal-place-search
Stars
182
Forks
17
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by brave on GitHub. Read the source before you install it.

Installation

Install the Local Place Search AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/brave/brave-search-skills/tree/main/skills/local-place-search
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/brave/brave-search-skills.git /tmp/brave-search-skills
mkdir -p .claude/skills
cp -r /tmp/brave-search-skills/skills/local-place-search .claude/skills/local-place-search
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Local Place Search in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Local Place Search on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Local Place Search is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Place Search (Search API)

Requires API Key: Get one at https://api.search.brave.com

Plan: Included in the Search plan (with locations option). See https://api-dashboard.search.brave.com/app/subscriptions/subscribe

Standalone: Unlike local-pois and local-descriptions, this endpoint does not require POI IDs from a prior web search. You provide a location and an optional query directly.

Quick Start (cURL)

Search by Query + Coordinates

bash
curl -s "https://api.search.brave.com/res/v1/local/place_search" \
  -H "Accept: application/json" \
  -H "Accept-Encoding: gzip" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}" \
  -G \
  --data-urlencode "q=coffee shops" \
  --data-urlencode "latitude=37.7749" \
  --data-urlencode "longitude=-122.4194" \
  --data-urlencode "radius=5000"

Search by Query + Location String

bash
curl -s "https://api.search.brave.com/res/v1/local/place_search" \
  -H "Accept: application/json" \
  -H "Accept-Encoding: gzip" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}" \
  -G \
  --data-urlencode "q=sushi restaurants" \
  --data-urlencode "location=tokyo japan" \
  --data-urlencode "country=JP" \
  --data-urlencode "search_lang=en"

Browse General POIs (No Query)

bash
curl -s "https://api.search.brave.com/res/v1/local/place_search" \
  -H "Accept: application/json" \
  -H "Accept-Encoding: gzip" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}" \
  -G \
  --data-urlencode "latitude=48.8566" \
  --data-urlencode "longitude=2.3522" \
  --data-urlencode "radius=3000" \
  --data-urlencode "country=FR"

Endpoint

http
GET https://api.search.brave.com/res/v1/local/place_search

Authentication: X-Subscription-Token: <API_KEY> header

Parameters

Location (optional but recommended)

Providing a geographic anchor improves precision. You can use coordinates (latitude + longitude) or a location string. Omitting both is allowed when a q is given — results are sourced globally and may be less precise. Omitting all three (q, latitude/longitude, and location) returns HTTP 422.

ParameterTypeRequiredDefaultDescription
latitudefloatConditional—Latitude (-90.0 to 90.0). Required together with longitude
longitudefloatConditional—Longitude (-180.0 to 180.0). Required together with latitude
locationstringNo—Location string, alternative to coordinates. US: <city> <state> <country> (e.g., san francisco ca united states). Non-US: <city> <country> (e.g., tokyo japan). Case-insensitive, no commas needed. English or the most popular local language works best

Search

ParameterTypeRequiredDefaultDescription
qstringNo—Free-text query (e.g., coffee shops, pizza). Fully optional — if omitted, returns general POIs in the given area

Additional Options

ParameterTypeRequiredDefaultDescription
radiusfloatNo—Search radius bias around the provided coordinates, in meters. Not a hard cutoff — results may extend beyond it. No upper limit
countintNo20Total items returned across all buckets (1–100), not just results — an address query can spend the whole budget on addresses/streets
geolocstringNo—User geolocation as <latitude>x<longitude> (e.g., 40.7128x-74.0060), used to compute distance
countrystringNoUSSearch country (2-letter country code or ALL)
search_langstringNoenLanguage for search results (2+ char language code)
ui_langstringNoen-USUI language (locale code, e.g., en-US)
unitsstringNometricMeasurement units: metric or imperial
safesearchstringNostrictSafe search level: off, moderate, or strict
spellcheckboolNotrueWhether to apply spellcheck to the query

Response Format

Top-Level Fields

FieldTypeDescription
typestringAlways "locations"
resultsarrayList of LocationResult objects (individual POIs)
citiesarrayMatched cities, type: "city" — see Geographic Place Fields
countriesarrayMatched countries, type: "country"
regionsarrayMatched regions, type: "region"
neighborhoodsarrayMatched neighborhoods, type: "neighborhood"
addressesarrayList of AddressResult objects with type: "address" — specific street + number locations
streetsarrayList of AddressResult objects with type: "street" — entire streets
mixedarrayResultReference ordering hints describing how to interleave the buckets on a SERP
locationobject?Resolved location info
location.coordinates[float, float][latitude, longitude] of the resolved center
location.namestringResolved location name (e.g., "Helsinki")
location.countrystringTwo-letter country code (e.g., "FI")

Treat a missing bucket as empty. For typical POI-style queries only results is populated, so clients that don't render rich SERPs can ignore the rest — except for address- or street-shaped queries, which can return an empty results and put every match in addresses/streets.

LocationResult Fields

Each item in results is a LocationResult:

FieldTypeDescription
typestringAlways "location_result"
titlestringBusiness/POI name
urlstringCanonical URL
descriptionstring?Short description or category label (e.g., "Coffee Shop")
provider_urlstringProvider page URL
idstring?Opaque POI identifier (valid ~8 hours, usable with local-pois and local-descriptions)
coordinates[float, float]?[latitude, longitude]
postal_addressobjectdisplayAddress, plus optional streetAddress, addressLocality, addressRegion, postalCode, country
contact.telephonestring?Phone number
contact.emailstring?Email address
rating.ratingValuefloat?Average rating
rating.bestRatingfloat?Max possible rating
rating.reviewCountint?Number of reviews
rating.is_tripadvisorboolWhether the rating comes from Tripadvisor
opening_hours.current_dayobject[]?Today's hours (abbr_name, full_name, opens, closes)
opening_hours.daysobject[][]?Hours for each day of the week
categoriesstring[]Business categories (default [])
price_rangestring?Price indicator, e.g. $, $$, $$ - $$$
serves_cuisinestring[]?Cuisine types (restaurants)
distance.valuefloat?Distance from the search location
distance.unitsstring?Distance unit
icon_categorystring?Icon category slug (e.g., cafe)
thumbnail.srcstring?Thumbnail image URL
thumbnail.originalstring?Original image URL
pictures.resultsobject[]?Additional images (src, original)
profilesobject[]?External profiles (name, url, long_name, img)
timezonestring?IANA timezone (e.g., America/Los_Angeles)
zoom_levelintSuggested map zoom level (default 7)

Geographic Place Fields (cities, countries, regions, neighborhoods)

All four buckets share one shape, differing only by the type identifier. The published spec names them CityResult / CountryResult / RegionResult / NeighborhoodResult.

FieldTypeDescription
typestringBucket identifier: city, country, region, or neighborhood
namestringPlace name
countrystringCountry code of the place
coordinates[float, float][latitude, longitude]
thumbnail.srcstringPrimary image URL

AddressResult Fields (addresses and streets)

Same model is used for both buckets. Items in addresses have type: "address" (street + number); items in streets have type: "street" (entire street).

FieldTypeDescription
typestring"address" (in addresses) or "street" (in streets)
namestringDisplay name of the address or street
coordinates[float, float][latitude, longitude]
poisobject[]LocationResult objects located at this address/street
pois_nearbyobject[]LocationResult objects located nearby
zoom_levelintSuggested map zoom level (default 15)
distance.valuefloat?Distance from the search location
distance.unitsstring?Distance unit
postal_addressobject?displayAddress, streetAddress, addressLocality, addressRegion, country

Mixed Ordering (mixed)

mixed is an ordered list of ResultReference objects telling clients how to interleave items from the different buckets on a single SERP.

FieldTypeDescription
typestringBucket to draw from: results, cities, countries, regions, neighborhoods, addresses, or streets
indexint?0-based index of the item within that bucket. May be null when all is true
allboolWhen true, all remaining items from the named bucket should be placed at this position

Clients that only render POIs can ignore mixed entirely and read results directly.

Example Response

json
{
  "type": "locations",
  "results": [
    {
      "type": "location_result",
      "title": "Blue Bottle Coffee",
      "url": "https://yelp.com/biz/blue-bottle-coffee-sf",
      "provider_url": "",
      "id": "loc4CQWMJWLD4VBEBZ62XQLJTGK6YCJEEJDNAAAAAAA=",
      "description": "Coffee Shop",
      "postal_address": {
        "type": "PostalAddress",
        "displayAddress": "315 Linden St, San Francisco, CA 94102"
      },
      "contact": { "telephone": "+15106533394" },
      "rating": {
        "ratingValue": 4.3,
        "bestRating": 5.0,
        "reviewCount": 1024,
        "is_tripadvisor": true
      },
      "opening_hours": {
        "current_day": [
          { "abbr_name": "Tue", "full_name": "Tuesday", "opens": "07:00", "closes": "18:00" }
        ],
        "days": [
          [{ "abbr_name": "Mon", "full_name": "Monday", "opens": "07:00", "closes": "18:00" }]
        ]
      },
      "coordinates": [37.7763, -122.4215],
      "categories": [],
      "serves_cuisine": ["Cafe", "Coffee Shop"],
      "price_range": "$$",
      "icon_category": "cafe",
      "thumbnail": {
        "src": "https://example.com/thumb.jpg",
        "original": "https://example.com/original.jpg"
      },
      "zoom_level": 7
    }
  ],
  "cities": [],
  "countries": [],
  "regions": [],
  "neighborhoods": [],
  "addresses": [],
  "streets": [],
  "mixed": [
    { "type": "results", "index": 0, "all": false }
  ],
  "location": {
    "coordinates": [37.7749, -122.4194],
    "name": "San Francisco",
    "country": "US"
  }
}

For a query that matches a city name, the response additionally surfaces a city entry in cities:

json
{
  "cities": [
    {
      "type": "city",
      "name": "San Francisco",
      "country": "US",
      "coordinates": [37.7749, -122.4194],
      "thumbnail": { "src": "https://example.com/sf.jpg" }
    }
  ],
  "mixed": [
    { "type": "cities", "index": 0, "all": false }
  ]
}

Enriching Results with POI Details and Descriptions

POI id values from results can be passed to sibling endpoints for richer data:

bash
# Get full POI details (hours, reviews, photos, web result mentions)
curl -s "https://api.search.brave.com/res/v1/local/pois" -G \
  --data-urlencode "ids=loc4CQWMJWLD4VBEBZ62XQLJTGK6YCJEEJDNAAAAAAA=" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}"

# Get AI-generated descriptions
curl -s "https://api.search.brave.com/res/v1/local/descriptions" -G \
  --data-urlencode "ids=loc4CQWMJWLD4VBEBZ62XQLJTGK6YCJEEJDNAAAAAAA=" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}"

Use Cases

  • Map-based exploration: Search for POIs within a visible map viewport using coordinates + radius. No prior query needed.
  • Location-aware apps: Build "nearby" features — pass device GPS coordinates and a query to find relevant businesses.
  • Travel planning: Search for attractions, restaurants, and hotels by location string (e.g., paris france) without needing exact coordinates.

Notes

  • Finds places, not pages: This endpoint searches a geographic index of physical places. Use web search for general information retrieval.
  • Choosing a radius: A tighter radius (below ~20 km) gives more focused results. Raise it to reach specific or well-known places further afield; for common category searches (e.g., restaurants), the default bias or tighter works best.

Frequently asked questions

What does the Local Place Search AI skill do?

USE FOR finding places in the physical world - businesses, POIs, street addresses, cities and streets. Results carry address, coordinates, rating, opening hours and phone, so basic details need no follow-up call. Standalone - no POI IDs or prior web search required; the IDs it returns work with local-pois and local-descriptions. Locate by coordinates or a location string, or omit both to search globally. Omit the query to browse an area. Max 100 results.

Why use Local Place Search on TypingMind?

Because you install it once and use it with any model. Local Place Search is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Local Place Search in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/brave/brave-search-skills/tree/main/skills/local-place-search. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Local Place Search?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Local Place Search?

As many as you like. As long as a model supports skills, you can use Local Place Search with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Local Place Search AI skill free?

Yes. It is published on GitHub by brave under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇