Skip to main content
MCP server CX Codex

kaiord — Claude Code MCP server: what it does, how to install

MCP server for parsing, validating, and converting workout files across FIT, TCX, ZWO, and KRD formats directly inside your coding agent.

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add kaiord -- npx -y @kaiord/mcp
Open on GitHub

What kaiord does

This server exposes the @kaiord/mcp tools to your coding agent via npx. It translates workouts between FIT, TCX, ZWO, and Kaiord's JSON-based KRD structure. Data checks rely on Zod schema validation.

Everything executes on your machine. The parser inspects interval steps by heart rate, wattage, duration, or calories. It extracts swimming pool dimensions, gear tags, and coaching instructions under 256 characters.

Be aware that the FIT adapter does not handle the training_peaks_tss interval duration yet. Node.js with npx is enough to run it.

What it helps with

  • convert cycling workout files from ZWO to FIT
  • validate TCX and KRD schemas with Zod
  • inspect target power and heart rate thresholds in interval blocks
  • extract swimming pool dimensions and drill notes from FIT files

Example prompts for the agent

  • Validate session.tcx and convert it into a FIT file
  • Inspect intervals.zwo and summarize power targets for each block
  • Export the swim_drill.krd file into TCX format

How to install kaiord

Platform CLI

claude mcp add kaiord -- npx -y @kaiord/mcp

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about kaiord

How to install kaiord in Claude Code?

Run claude mcp add kaiord -- npx -y @kaiord/mcp in your terminal. This registers the server and runs the package directly via npx.

What formats does kaiord handle?

It converts and validates workout structures across FIT, TCX, ZWO, and the native JSON-based KRD format.

What limitations does the FIT converter have?

Step instructions cannot exceed 256 characters. Also, duration intervals based on training_peaks_tss are currently not implemented.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor