Skip to main content
Glama
Mountaineers MCP Server

Mountaineers MCP Server

by dreamiurg

Mountaineers MCP Server

npm

Other projects you might like: PNW Climb Planner · mountaineers-assistant · peakbagger-cli · claude-mountaineering-skills

What is this?

This lets AI assistants like Claude and ChatGPT search and browse mountaineers.org on your behalf. Instead of clicking through the website, you can ask questions in plain English:

  • "Find me a beginner-friendly day hike near Seattle this weekend"

  • "What scrambles are available in August?"

  • "Show me trip reports for Mount Pilchuck"

  • "What's the elevation gain on the Mount Si trail?"

  • "What activities am I signed up for?"

  • "What badges have I earned?"

The AI reads the Mountaineers website, understands the results, and gives you a conversational answer -- no manual searching required. Read the full story: I Built an MCP Server for Mountaineers.org.

Related MCP server: MiniMax-Coding-Plan-MCP

What can it do?

Search public data:

  • Search activities by type, branch, difficulty, date, and more

  • Search courses, clinics, and seminars

  • Browse trip reports

  • Search routes and places

  • Get full details for any activity, trip report, route, or course

Access your account:

  • See your upcoming and past activities

  • See your completed activity history

  • See your course enrollments

  • View your earned badges and certifications

  • View member profiles and activity rosters

Setup

IMPORTANT

A one-time login is required for all tools, including public search. mountaineers.org is behind Cloudflare, which blocks plain HTTP clients. After installing the server (below), just ask your assistant to run the login tool — a Chrome window opens, you sign in once on mountaineers.org, and the server caches the clearance + session cookies. Your password is typed into the real site, never into the chat or any config. See Authentication for details and the requirements (desktop with Google Chrome).

Follow the instructions for your AI app below.

Claude Desktop

  1. Download mountaineers-mcp-X.Y.Z.mcpb from the latest release — one bundle covers Intel and Apple Silicon Macs and Windows

  2. Open Claude Desktop → Settings → Extensions → Install Extension

  3. Select the downloaded .mcpb file

  4. Ask your assistant to run the login tool (one time) — a Chrome window opens, you sign in, and authentication is cached. Required before any other tool works. See Authentication.

That's it -- no Node.js setup required. Just run the one-time Authentication step (the login tool) before using other tools.

Requires Node.js 18+.

  1. Go to Settings > Developer > Edit Config

  2. Paste this and save:

{
  "mcpServers": {
    "mountaineers": {
      "command": "npx",
      "args": ["-y", "mountaineers-mcp"]
    }
  }
}
  1. Quit and reopen Claude Desktop (not just close the window -- fully quit)

Then run the Authentication step (the login tool). Don't add MOUNTAINEERS_USERNAME/MOUNTAINEERS_PASSWORD to this config — the server reads the cached cookie, not env vars.

ChatGPT Desktop

ChatGPT Desktop supports MCP through Developer Mode, but only remote HTTP servers -- it cannot run local command-line tools like Claude Desktop can. To use this server with ChatGPT Desktop, you would need to run it behind a tunnel (e.g., mcp.run or ngrok). This is not yet streamlined; we plan to add Streamable HTTP transport in a future release.

Requires ChatGPT Plus, Pro, Team, or Enterprise.

Claude Code (CLI)

Run this in your terminal:

claude mcp add mountaineers -- npx -y mountaineers-mcp

Or add to your .mcp.json:

{
  "mcpServers": {
    "mountaineers": {
      "command": "npx",
      "args": ["-y", "mountaineers-mcp"]
    }
  }
}

Then run the Authentication step (the login tool).

Codex CLI

Add to ~/.codex/config.toml:

[mcp_servers.mountaineers]
command = "npx"
args = ["-y", "mountaineers-mcp"]

Then run the Authentication step (the login tool).

Or use the CLI:

codex mcp add mountaineers -- npx -y mountaineers-mcp

Authentication

Mountaineers.org is behind Cloudflare, which blocks plain HTTP clients — so every tool (public search included) needs a valid Cloudflare clearance cookie, plus a session cookie for account tools. The server doesn't log in itself; it replays cookies you mint once with a real browser.

Just ask your assistant to run the login tool (e.g. "log in to mountaineers"). A Chrome window opens to the mountaineers.org sign-in page; you complete the login there (your password goes to the site, never to the chat, the server, or any config). The server then caches the cookies and every other tool works. If the browser is already signed in from a previous run, you won't need to type anything (it still takes a few tens of seconds to open Chrome, settle the session, and close).

The cookies are saved to ~/.cache/mountaineers-mcp/clearance.json (or under $XDG_CACHE_HOME), owner-readable only. The server reads the cache on startup and re-reads it automatically if the clearance expires mid-session — so re-running login takes effect without a restart.

Requirements: a desktop with Google Chrome installed and a visible display. (Headless servers can't show the sign-in window.)

When to re-run: Cloudflare clearance eventually expires (lifetime is set by Cloudflare and varies). When a tool reports the clearance expired, run login again.

Alternative: npm run login (from a checkout)

For scripted/automated setups you can mint the cache from a local checkout, auto-filling credentials from the environment:

git clone https://github.com/dreamiurg/mountaineers-mcp.git
cd mountaineers-mcp
npm install
MOUNTAINEERS_USERNAME=your-username MOUNTAINEERS_PASSWORD=your-password npm run login

This writes the same cache file the server reads. The credentials are used only by this command (passed as env vars, not persisted).

Tools reference

Run login once before anything else (Cloudflare gates the whole site); the data tools below then work, split by the kind of data they return.

Setup

Tool

Description

login

Open a browser to sign in to mountaineers.org and cache the Cloudflare clearance + session cookies. Required once before any other tool; re-run when clearance expires.

Public data

Tool

Description

search_activities

Search activities with filters (type, branch, difficulty, date, day of week)

search_courses

Search courses, clinics, and seminars

search_trip_reports

Search trip reports by text and activity type

search_routes

Search routes and places with filters (activity type, difficulty, climbing category)

get_activity

Get full activity details (leader notes, route, equipment)

get_trip_report

Get trip report details (conditions, route info)

get_route

Get route details (difficulty, elevation, directions, maps, related routes)

get_course

Get course details (schedule, pricing, leaders, badges earned)

Your account

Tool

Description

whoami

Get your name, profile URL, and member slug

get_my_activities

Your registered activities (upcoming) with filtering

get_my_courses

Your course enrollments with filtering

get_activity_history

Your completed activity history with filtering by result, type, and date

get_my_badges

Your earned badges and certifications with dates

get_member_profile

View a member's profile, badges, and committees

get_activity_roster

See who's signed up for an activity

Privacy

Your credentials are used only by npm run login (passed as environment variables, not persisted) and are sent only to mountaineers.org. What's saved on your computer is the resulting session cookie cache (~/.cache/mountaineers-mcp/clearance.json, owner-readable only). Nothing — credentials or cookies — is ever sent to any AI provider or third party.

Development

npm install
npm run dev          # Run from TypeScript sources via tsx (no build step)
npm run check        # Typecheck + lint
npm test             # Run tests
npm run ci           # Full CI: check + coverage + build

License

MIT


More from @dreamiurg

Made by @dreamiurg in Seattle. If this project saved you time, you can buy me a coffee — appreciated, never expected.

Available Tools

28 tools
get_activityB

Get detailed information about a specific activity by URL or slug.

Inputs: url (required).

get_activity_historyA

Get the logged-in user's completed activity history (past trips, courses, events). Supports filtering by category, result, activity type, and date range. For arbitrary members use get_member_history.

Inputs: limit, result, date_to, category, +2 more.

get_activity_rosterA

Get the participant roster for a specific activity by URL or slug. Requires authentication.

Inputs: url (required).

get_badgeA

Get detailed information about a specific mountaineers.org badge by URL: title, description, badge_type, branch_slug (for branch-scoped award badges), body text with criteria/instructions, and an optional categories array (for award badges with Category A/B/C/... requirements). Earners list is not exposed by the site and is not returned. Use search_badges to discover badge URLs; for the authenticated user's earned badges use get_my_badges.

Inputs: url (required).

get_courseB

Get detailed information about a specific course, including schedule, pricing, leaders, and badges earned.

Inputs: url (required).

get_eventA

Get detailed information about a specific mountaineers.org event by URL: title, description, date/time string, committee, branch, body text, and any additional event-specific fields (expected attendance, contacts, custom Q&A).

Inputs: url (required).

get_member_activitiesA

Get a member's upcoming/registered activities by slug or profile URL. Returns activities the member is currently signed up for — for past completed activities use get_member_history. Mirrors get_my_activities for any member. Requires authentication.

Inputs: member (required), limit, result, status, +3 more.

get_member_coursesA

Get a member's courses (upcoming + past, merged into one list ordered by enrolled date ascending), given their slug or profile URL. Mirrors get_my_courses for any member. Requires authentication.

Inputs: member (required), role, limit, result, +3 more.

get_member_historyA

Get the past activity history for any mountaineers.org member, given their slug or profile URL. Returns completed activities only (not future-registered). Same filters as get_activity_history (which is the self-only version). Requires authentication.

Inputs: member (required), limit, result, date_to, +3 more.

get_member_profileA

Get a member's profile information including name, branch, committees, and badges. Requires authentication.

Inputs: member_slug (required).

get_my_activitiesA

Get the logged-in user's registered activities (upcoming and past). Supports filtering by category, status, result, and date range.

Inputs: limit, result, status, date_to, +2 more.

get_my_badgesA

Get the logged-in user's earned badges and certifications. Supports filtering by name and active status.

Inputs: name, active_only.

get_my_coursesA

Get the logged-in user's course enrollments (current and past). Supports filtering by status, role, result, and date range.

Inputs: role, limit, result, status, +2 more.

get_routeA

Get detailed information about a specific route or place on mountaineers.org, including difficulty, elevation, directions, and maps.

Inputs: url (required).

get_route_trip_reportsA

List trip reports filed against a specific route/place URL. Paginated, 20 per page; results follow the site's default order (typically newest first). Use this when you have a route URL and want member-written reports for that route specifically — more accurate than searching trip reports by route name.

Inputs: route_url (required), page.

get_trip_reportA

Get detailed information about a specific trip report by URL or slug.

Inputs: url (required).

get_versionA

Get the mountaineers-mcp server name and version. Useful for verifying which server build a client is connected to when debugging version-specific behavior.

Inputs: no named parameters.

list_branchesA

List all mountaineers.org branches (e.g. Seattle, Tacoma, Olympia). Returns slug + display name + URL. Use the slug as input to list_committees.

Inputs: no named parameters.

list_committeesA

List all committees for a given branch (e.g. 'seattle-branch'). Returns slug + name + URL.

Inputs: branch (required).

loginA

Sign in to mountaineers.org. Opens a Chrome window where YOU complete the login (your password is never sent to this server or the AI); it then caches your session. Required once before any other tool works — mountaineers.org blocks unauthenticated clients. Re-run when a tool reports your session has expired. Requires a desktop with Google Chrome (stable channel) installed; the call blocks for up to ~3 minutes while you sign in. If you're already signed in from a previous run, it may finish in seconds.

Inputs: no named parameters.

search_activitiesA

Search currently-published outdoor activities on mountaineers.org (upcoming trips, course field trips, clinics). Supports filtering by type, branch, difficulty, date range, and more. Past activity instances are NOT in the search index — if you have a known historical activity URL use get_activity, or use get_member_history for a member's past activities.

Inputs: page, type, query, branch, +7 more.

search_badgesB

Search the mountaineers.org badge catalog. Returns title, URL, and badge_type derived from the URL (one of: award, instructor, leader, other). Optional free-text query.

Inputs: page, query.

search_coursesA

Search currently-published courses, clinics, and seminars on mountaineers.org (upcoming and rolling-enrollment only). Past course instances (prior years) are NOT in the search index, even though their URLs still resolve — if you have a known historical course URL use get_course, or use get_my_courses for the authenticated user's past courses.

Inputs: page, query, branch, open_only, +2 more.

search_eventsA

Search mountaineers.org events (festivals, clinics, social gatherings, branch meetings — distinct from activities and courses). Returns title, URL, date string, and location. Optionally filter by free-text query.

Inputs: page, query.

search_membersA

Search the mountaineers.org member directory by name or text. Returns name + slug + profile URL. The directory includes both individual members and group/role accounts (e.g. committee mailboxes); both are returned with no type marker. Requires authentication and a non-empty query.

Inputs: query (required), page.

search_routesB

Search routes and places on mountaineers.org. Filter by activity type, difficulty, climbing category, and more.

Inputs: page, query, used_for, difficulty, +3 more.

search_trip_reportsA

Search trip reports on mountaineers.org. Trip reports are member-written accounts of past outings; the full historical archive (8000+ reports) is searchable here. To narrow to a specific route, pass the route name as the query.

Inputs: page, query, activity_type.

whoamiA

Get the currently logged-in user's name, slug, and profile URL. Requires an active session — run the login tool first.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 28 tools

    Disambiguation5/5

    Each tool targets a distinct resource and action: search_* for discovery, get_* for detail, and get_my_*/get_member_* clearly separate self vs. other-member data. The only near-overlap, search_trip_reports vs. get_route_trip_reports, is explicitly disambiguated in the descriptions.

    Naming Consistency4/5

    The vast majority of tools follow a consistent verb_noun snake_case pattern (search_*, get_*, list_*), with clear my_ and member_ prefixes. Minor exceptions are login and whoami, which are standalone commands outside the verb_noun pattern.

    Tool Count2/5

    With 28 tools, this is above the 25-tool threshold that indicates an oversized surface. Many of the get_my_*/get_member_*/history variants could potentially be consolidated, making the set heavier than necessary even though each tool is individually clear.

    Completeness4/5

    The toolset covers discovery and detail for all major content types—activities, courses, events, trip reports, routes, and badges—plus authentication, branches, committees, and both self and member-specific views. Minor gaps exist: there is no member-specific event listing beyond activity history, and committee data stops at listing without deeper detail.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      A
      maintenance
      Provides tools for AI-powered web search and image understanding, returning organic search results or analyzing images from URLs or local files based on text prompts. Integrates with MCP clients such as Claude Desktop, Cursor, and Windsurf to support development workflows.
      -
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables AI agents to search the web across configured providers and crawl single pages or entire sites into clean Markdown or JSON. Supports deduplication, link stripping, and authenticated crawls behind a login.
      -
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables AI agents to search the web through configured providers and crawl single pages or whole sites into clean Markdown or JSON, including content behind logins, over MCP STDIO or HTTP.
      -