How to use Claude to improve shadow-aware routing
An author used Claude Code to clean up a routing app built with Antigravity, then live-tested its shadow-aware routes and identified missing tree data.
news.ycombinator.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Help clean up and improve [existing routing app]. Compare [base-route method] with A* shadow-aware routing using [geographic data]. Show sunlight exposure and travel-time differences, add debugging overlays for validation, and flag data gaps for live testing.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- TeMPOraL
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author said the app mostly matched reality in live testing and worked within the limits of available OSM data, with missing tree data a major gap.
- Human involvement
- The author used debugging overlays to validate and diagnose the app and tested it live, finding a gap in tree data.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Built the routing app while trying Antigravity.
- Cleaned up and improved the app with Claude Code.
- Tested the app live against real-world conditions.
TeMPOraL built a routing application while trying Antigravity, then cleaned up and improved it with Claude Code. The app compared an ordinary route between two map points with a route that accounted for shadows. It did not yet support address lookups.
For the base path, the application used either local A* or an OSMR endpoint, then used A* for shadow-aware routing. It displayed both routes alongside statistics about total exposure time, sunlight exposure saved and additional travel time relative to the base route. Debugging overlays helped the author validate and diagnose pathfinding and data quality.
The author tested the app live and said it mostly matched reality. A major gap was missing tree data. They considered the app to be doing its job about as well as available OSM data allowed and were looking at improving that data next.
They withheld a public link because the first page load served 12MB of geo JSON to the client and they were concerned about a traffic bill. Who it's for: developers exploring route comparisons who can validate geographic data and routing behavior outdoors.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · TeMPOraL
Frequently asked questions
How did the app compare routes?
It showed a base route and a shadow-aware route, with statistics for exposure time, sunlight exposure saved and additional travel time.
What did live testing reveal?
The author said the app mostly checked out against reality, but missing tree data was a major gap.
More details
Remaining steps
- Identified missing tree data and looked at improving the data.
About this account
The author used Claude Code for cleanup and improvements but did not specify which changes it made or how independently it worked.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=48262424. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: TeMPOraL on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
Claude Code · Antigravity · A* · OSM · OSMR endpoint
Routing · Maps · OSM
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.