bikescout
AI for tactical cycling intelligence: Predictive mud logic, S-Scale terrain analysis, and powered GPX routing
Documentation
BikeScout
Open-source AI cycling route planner for MTB, gravel, road and e-bike riders
BikeScout is a specialized AI engine for MTB, Road, E-Bike, and Gravel mission planning.
It transforms raw map data into Tactical Intelligence, predicting terrain conditions and trail hazards.
The system provides precise setup advice, tailoring your equipment to the demands of the specific route, identifying technical challenges and environmental risks before you even leave the garage.
Quickstart
How to use BikeScout
- instruction, using the **BikeScout GUI**
- instruction, using your favourite MCP Agent client: Claude Desktop, Cursor, Codex, Google Antigravity, Windsurf
๐ฅ Click to watch BikeScout GUI
๐ฅ Click to watch a Demo
Explore BikeScout
- **User Guide** Detailed API references and step-by-step guides to deploy the BikeScout.
- **Website** Explore the visual breakdown of Mud Logic, S-Scale terrain grading, and how we transform raw OSM data into mission-ready intel. Check the latest race analysis reports from the legendary routes.
Love BikeScout? โญ Star this repo to support the development of the first open-source tactical cycling engine.
Found a bug? Open an Issue. Want to add a local skill? PRs are welcome!
Example Queries
You can ask BikeScout complex, multi-step requests. It combines real-time data with technical cycling intelligence to provide expert-level answers.
๐บ๏ธ Advanced Planning (Multi-Tool)
- *"I'm at Monte Cavo with my MTB bike (29 tires). Plan a 25km loop for me. Check if the terrain is compatible with my bike, verify the afternoon rain probability, and suggest a 'Fraschetta' for the finish. Use the Castelli Romani guide."*
โ๏ธ Bike Setup & Surface Intelligence
- *"Check this route from Barcelona city center to El Prat. I'm on a Road Bike with 25mm tires. Is it compatible? Give me the exact percentage of gravel vs asphalt."*
๐ Visual Elevation & Gradient Analysis
- *"Plan a 40km route starting from Bormio. I need the Visual Elevation Profile to see the exact gradients of the Stelvio climb. Highlight sections over 12% so I can manage my pacing."*
๐๏ธ Local Expertise
- *"Use the Derby local guide to plan a road cycling route starting from Derby. I need at least 800m of elevation gain. Also, recommend the correct tire pressure for high-altitude descents."*
๐ ๏ธ Quick Tech Checks
- *"Give me the safety checklist and calculate the tire pressure for a 90kg rider on 2.3" tubeless tires for a muddy ride."*
๐ Post-Ride Analysis & Terrain Truth
- *"Analyze my ride from 2026-04-12. Compare my average speed with the Mud Risk at that time and tell me if the terrain conditions were the reason for my slow pace."*
Why BikeScout?
While others products focus on where to go, BikeScout focuses on how you can thrive on that specific ride, right now.
| Platform | What they provide | Why BikeScout is superior |
|---|---|---|
| Komoot | Static maps | We add live risk-context |
| Strava | History/Social | We add predictive safety |
| Trailforks | Databases | We add personalized intelligence |
| Generic AI | Theory | We add real-world telemetry |
๐ค Contributing
Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make to BikeScout are greatly appreciated.
How to Contribute
1. Report Bugs: Found a glitch? Open an Issue with a detailed description and steps to reproduce.
2. Feature Requests: Have an idea to make BikeScout better? Open an issue to discuss it!
3. Pull Requests:
Coding Standards
- Please follow PEP 8 for Python code.
- Ensure all new tools are documented in the `README.md`.
- Keep comments in English for international collaboration.
*By contributing, you agree that your contributions will be licensed under the project's AGPLv3 License.*
License & Data Attributions
Software License
This project is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0) - see the LICENSE file for details.
Why AGPLv3?
BikeScout contains proprietary-grade tactical logic (such as the TAELยฎ Mud Reservoir Model). The AGPLv3 ensures that:
- Transparency: Any modified version of BikeScout used to provide a service over a network (SaaS/Cloud) must make its full source code available to the community.
- Integrity: The core tactical intelligence remains open and collaborative, preventing "closed-source" commercial hijacking of the platform's unique algorithms.
Data Sources & Credits
BikeScout aggregates data from several open providers. Users of this server must adhere to their respective terms:
- Routing & Map Data: Provided by OpenRouteService by HeiGIT.
- Geospatial & Geocoding Data: ยฉ OpenStreetMap contributors. Data is available under the Open Database License (ODbL). Geocoding service powered by Nominatim.
- Weather Forecasts: Powered by Open-Meteo. Data is licensed under CC BY 4.0.
Frequently asked questions
What is bikescout?
bikescout is AI for tactical cycling intelligence: Predictive mud logic, S-Scale terrain analysis, and powered GPX routing
How do I install bikescout?
Open the GitHub repository and follow its README. Most MCP servers are added to your client's MCP config, then called by your agent.
Is bikescout open source?
Yes โ it is hosted on GitHub at https://github.com/hifly81/bikescout and has 20 stars.
Related MCP tools
Open-source coding agent memory. Records issues, attempts, fixes and decisions, then warns your agent before it repeats an approach that already failed. Native MCP server for Claude Code, Cursor, Antigravity and Codex. 100% local, no cloud, no telemetry. MIT.
An AI Gateway, registry, and proxy that sits in front of any MCP, A2A, or REST/gRPC APIs, exposing a unified endpoint with centralized discovery, guardrails and management. Optimizes Agent & Tool calling, and supports plugins.
AI-powered OSINT agent with interactive REPL, MCP server, and CLI. 19 tools. Works with Claude, GPT-4, or local models. For authorized security research only.
Cognee is the open-source AI memory platform for agents. Give your AI agents persistent long-term memory across sessions with a self-hosted knowledge graph engine.
Build effective agents using Model Context Protocol and simple workflow patterns Python-based implementation. Trusted by 7600+ developers.
โก ้ๆฑๅๆๆ็ๆๅ 200%๏ผๅ จ็้ฆไธชไธบ AI ็ผ็จๆถไปฃ่ฎพ่ฎก็ๅข้ๅไฝ MCP ๆๅกๅจ๏ผ่ชๅจๅๆ้ๆฑ่ชๅจ็ผๅๅๅ็ซฏไปฃ็ ๏ผไธ่ฝฝๅๅพ
Run your own MCP server? See who uses it and what to fix.
Measure it with TrackMCP