AllTrails MCP Server
Pricing
Pay per event + usage
AllTrails MCP Server
MCP server + scraper for AllTrails. Search hiking trails, fetch full details, reviews and weather live from Claude, ChatGPT, Cursor and other AI agents — or run it as a classic scraper.
Pricing
Pay per event + usage
Rating
0.0
(0)
Developer
axly
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
9 hours ago
Last modified
Categories
Share
Give your AI assistant live access to AllTrails hiking data. This actor runs as a Model Context Protocol (MCP) server so agents like Claude, ChatGPT, Cursor, and n8n can search trails, pull full trail details, read reviews, check weather, and autocomplete places — on demand. It also runs as a classic scraper that pushes trail rows to a dataset.
Why MCP
Instead of pre-scraping and storing data, your agent queries AllTrails live while it reasons — perfect for trip-planning assistants, outdoor chatbots, and research copilots. No other AllTrails MCP server exists on the Store.
Tools exposed
| Tool | What it does |
|---|---|
search_trails | Search trails by place name or lat/lng — length, elevation, difficulty, GPS, rating, activities |
get_trail_details | Full description, highlights, tips, getting-there, parking, points of interest, rating breakdown |
get_trail_reviews | Recent user reviews for a trail (sortable) |
get_trail_weather | Multi-day weather forecast for a trail |
suggest_places | Autocomplete places, parks and trails for a partial query |
Who it's for
- AI app builders — add trail search + detail to an assistant with a few MCP tools.
- Trip-planning bots — combine
search_trails+get_trail_weatherfor weather-aware suggestions. - Outdoor content copilots — pull descriptions and reviews on demand.
Connect your agent
Run the actor in Standby mode and point your MCP client at the server URL:
https://<your-actor-standby-url>/mcp
Example (Claude Desktop mcp config):
{"mcpServers": {"alltrails": {"url": "https://<your-actor-standby-url>/mcp","headers": { "Authorization": "Bearer <APIFY_TOKEN>" }}}}
Then ask: "Find moderate trails near Lake Tahoe with great views, and check the weather for the top one."
Classic scraper mode
Run it normally with an input (place, coordinates, filters) and it pushes trail rows to the dataset — same fields as the AllTrails Trail Scraper.
Output fields (dataset mode)
| Field | Type | Description |
|---|---|---|
id | int | Trail ID |
name | string | Trail name |
url | string | AllTrails URL |
latitude / longitude | number | Trailhead GPS |
areaName / stateName / countryName | string | Location |
lengthMiles | number | Length |
elevationGainMeters | number | Elevation gain |
difficulty | string | easy / moderate / hard |
avgRating / numReviews | number/int | Community rating |
activities / features | array | Tags |
scrapedAt | string | Timestamp |
Example input (classic run)
{ "query": "Yosemite National Park", "maxItems": 10, "sort": "most_popular" }
FAQ
Do I need a login or proxy? No — the actor handles AllTrails' anti-bot automatically and needs no account.
How fresh is the data? Every tool call fetches live data.
What data source is used? AllTrails' public mobile API — the same data the app shows.
Is scraping legal? You are responsible for complying with AllTrails' terms and applicable laws. Use the data responsibly.