---
title: "hills"
description: "Tiny hills above the prompt, one per metric a session is hillclimbing, showing how far up each is; hover one for details and an outside read on whether there is more to climb"
canonical: https://agentpluginsdirectory.com/plugins/hills
last-updated: 2026-10-05
---

# hills
Tiny hills above the prompt, one per metric a session is hillclimbing, showing how far up each is; hover one for details and an outside read on whether there is more to climb
- Slug: hills
- Publisher: Scott Smerchek
- Repository: https://github.com/smerchek/agent-plugins
- Manifest: plugins/hills/plugin.json
- Version: 0.2.0
- License: MIT
- Category (editorial): other
- Skills: 0
- MCP servers: 0
- Stars: 0
- Repository created: 2026-10-03
- Repository last pushed: 2026-10-03
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/hills
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What hills does, in the publisher's words

A Claude Code mod that shows a hillclimb as a row of tiny hills above the prompt, one per metric. The climbed part of each hill is solid green and the rest is a grey outline, so you can see at a glance how far up each one is. A faint blue rise after a hill means there is more to climb past it. Each hill has a short label (img, boot, cov). Hover a hill for its details: baseline to now, target or estimated ceiling, rounds tried and reverted, and the outside read.

It needs no log file or format. After each turn it forks the session's own transcript ($.model.fork, served mostly from the prompt cache) and asks which metrics are being climbed, with their baseline, rounds, reverts and any stated target.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/smerchek/agent-plugins/HEAD/plugins/hills/README.md