hevy-mcp
Unofficial MCP server for Hevy that lets AI assistants design training programs, create routines directly in your Hevy account, and analyze logged workouts.
README
hevy-mcp
Unofficial MCP server for Hevy. Design training programs in chat; your AI creates the routines directly in your Hevy account and analyzes your logged training. Requires a Hevy PRO API key (hevy.com/settings?developer).
Use it
Hosted (easiest — claude.ai web, mobile, desktop)
Add a custom connector in claude.ai → Settings → Connectors:
https://hevy-coach.ajwallacemusic.workers.dev/mcp
You'll be prompted for your Hevy API key; it's stored encrypted in your own OAuth grant and sent only to api.hevyapp.com.
Claude Desktop one-click
Download hevy-mcp.mcpb,
double-click, paste your key when prompted.
Local (Claude Code / Cursor / any stdio client)
claude mcp add hevy -e HEVY_API_KEY=<key> -- npx -y @almostjacked/hevy-mcp
Self-host (free, your own infra)
The published package ships the worker — no clone needed:
mkdir hevy-worker && cd hevy-worker && npm init -y
npm install @almostjacked/hevy-mcp wrangler
cat > wrangler.jsonc <<'EOF'
{
"name": "hevy-mcp",
"main": "node_modules/@almostjacked/hevy-mcp/dist/worker.js",
"compatibility_date": "2026-07-01",
"compatibility_flags": ["nodejs_compat"]
}
EOF
npx wrangler deploy
npx wrangler secret put AUTH_TOKEN # any long random string — becomes your Bearer token
npx wrangler secret put HEVY_API_KEY # from hevy.com/settings?developer
Add your worker URL (https://hevy-mcp.<your-subdomain>.workers.dev/mcp) to clients
that support custom headers — e.g. Claude Code:
claude mcp add --transport http hevy-selfhost <url> --header "Authorization: Bearer <token>".
claude.ai custom connectors don't support custom auth headers; for claude.ai use the
hosted instance above, or front your worker with Cloudflare Access.
Tools
| Tool | Description |
|---|---|
| search_exercises | Search the user's Hevy exercise library by name (fuzzy). Returns template id, exact title, type, muscle group and equipment. |
| list_routine_folders | List the user's routine folders (id + title). Check this before creating a folder — reuse an existing one when the name matches. |
| create_routine_folder | Create a routine folder, or return the existing folder if one with this exact title already exists (never duplicates). |
| list_routines | List the user's existing routines (id, title, folder_id, exercise_count). Use to avoid creating duplicates. |
| create_routine | create a routine (lb-first: weight_lb snaps to 2.5/5 lb increments; RPE folded into notes; supersets; duration/distance targets supported) |
| get_workouts | Get the user's most recent logged workouts, newest first, as compact summaries: title, date, duration, and per-exercise set count + top set. Use get_workout for full set-by-set detail of one workout. |
| get_workout | Get one logged workout with every set (weights in kg). Use the id from get_workouts. |
| get_workout_count | Total number of workouts the user has ever logged in Hevy. |
| get_exercise_history | past sets + estimated 1RM in lb for one exercise (full history, optional date range) |
| get_routine | Get one routine in full (all exercises and sets). Always call this before update_routine so you can send back the complete routine. |
| update_routine | Replace an existing routine. WARNING: this overwrites the routine — Hevy replaces its exercises with exactly what you send, so first call get_routine and include every exercise you want to keep. Folder cannot be changed here; duration/distance/custom-metric set targets round-trip safely. |
| get_training_summary | weekly sessions/volume (lb), sets per muscle group, top-lift 1RM trend |
Privacy Policy
Your Hevy API key and training data pass through only to api.hevyapp.com. The stdio/.mcpb versions run entirely on your machine. The hosted instance stores nothing but your encrypted OAuth grant; disconnecting revokes it. Nothing is logged or shared with third parties.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.