bhoudebert

Agent Motoride — Development skill for Claude Code

Development community

Agentic motorcycle ride planner: one sentence in, a ride you can follow out.

How to install Agent Motoride

This entry records only its repository, not the path inside it, so there is no exact command to give. Open bhoudebert/agent-motoride and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Motoride does

Agentic motorcycle ride planner: one sentence in, a ride you can follow out. Claude tool-use loop with scouts, deterministic road/weather/stop tools, ride library, MCP server for Claude Code and Codex.

Alternatives in Development

  • Planner 24.1k ★
  • Tentacle Planner — You are the Tentacle Planner — a meta-agent that analyzes this codebase and creates department tentacles to or 1.4k ★
  • /autopilot — Autonomous hunt loop with deterministic scope safety and configurable checkpoints 1.2k ★

README

agentRide

[![CI](https://github.com/bhoudebert/agent-motoride/actions/workflows/ci.yml/badge.svg)](https://github.com/bhoudebert/agent-motoride/actions/workflows/ci.yml) [![CodeQL](https://github.com/bhoudebert/agent-motoride/actions/workflows/codeql.yml/badge.svg)](https://github.com/bhoudebert/agent-motoride/actions/workflows/codeql.yml) [![Release](https://img.shields.io/github/v/release/bhoudebert/agent-motoride?display_name=tag&sort=semver)](https://github.com/bhoudebert/agent-motoride/releases) [![Node 24](https://img.shields.io/badge/node-%E2%89%A524-339933?logo=node.js&logoColor=white)](.nvmrc) [![Licence MIT](https://img.shields.io/badge/licence-MIT-blue.svg)](LICENSE) [![Conventional Commits](https://img.shields.io/badge/commits-conventional-fe5196?logo=conventionalcommits&logoColor=white)](CONTRIBUTING.md)

A motorcycle ride planner driven by an AI agent. You say what you want in one sentence:

This Saturday, no rain, under 250 km, winding roads, give me an itinerary

and you get a ride you can follow: the loop, the roads, estimated riding time, how much of it is open road, the forecast along the way, sunrise and sunset, fixed speed cameras, where to fuel and where to stop for coffee, and the links and files to put it on your phone.

The agent decides where to look and what to propose. Everything it states comes from tools: road geometry and speed limits from OpenStreetMap, routing from Valhalla, forecasts from Open-Meteo, traffic from TomTom when you have a key. The code enforces your hard rules (no motorways unless you say so, a saved ride is never silently duplicated) and keeps a library of rides you liked, with ratings, so the next ride is different and better.

It is also a working example of an agentic application: tool use, parallel sub-agents ("scouts"), schema-validated answers, a replayable trace of every step, cost accounting and a benchmark of models, and the same tools exposed over the Model Context Protocol.

![How agentRide works: the