HN Showcase
Back to browse
GitHub Repository

Cursor extension that tracks the prompt cache TTL for your LLM chat sessions.

3 starsTypeScript

Cursor extension to track LLM cache TTL

by agastalver·Apr 3, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidNiche GemSolve My Problem

Finally shows you when your Cursor cache expires before you lose money.

Strengths
  • •Status bar countdown per chat with color-coded urgency (green/yellow/red)
  • •Auto-detection watches Cursor's agent transcript files for activity changes
  • •Cache Keep feature nudges you to send keep-alive messages before TTL expires
Weaknesses
  • •WSL projects don't work correctly — timers and chat detection break
  • •Narrow audience: only matters if you use Cursor and care about cache costs
Target Audience

Cursor/VS Code users leveraging LLM features with caching

Post Description

Hey HN,

I built a small Cursor extension that tracks LLM cache timers so you can see when your context cache is about to expire.

While working at Ploid AI (ploid.ai), I noticed that most cache rewrites, both mine and my teammates', came down to two things: forgetting to continue the right chat, or forgetting to review plans before the cache dropped off. Basically, money and context lost to simple timing mistakes.

So I built this extension to give that extra information. It won't magically make you a better coder or promise cost savings, but it gives you a bit more visibility into what's happening under the hood.

GitHub: https://github.com/agastalver/cache-timer-extension

Extension: https://open-vsx.org/extension/agastalver/cache-timer-extens...

Would love feedback or suggestions.

Similar Projects