Show HN: Chrome Extension Claude Token Usage Bar and Context Use for Claude.ai
**Show HN** listed a **Chrome extension** that adds a **Claude token usage bar** and **context-use** display on **Claude.ai**. The project lives at…
By Dillip Chowdary • Aug 03, 2026 • Source: HN Claude/Codex/Fable
**Show HN** listed a **Chrome extension** that adds a **Claude token usage bar** and **context-use** display on **Claude.ai**. The project lives at **github.com/disi910/claude-usage-bar**. On Hacker News it sat at **2 points** with **0 comments** under item **49010415**.
The product is a browser extension aimed at the Claude.ai web UI, not a Claude API client or desktop app. Its surface is a visible **token usage bar** plus a **context use** readout while you work in chat. That implies the extension reads session or page-level usage signals available in the browser and renders them continuously so the user does not have to leave the conversation or open a separate billing page.
Advertisement
Tech Pulse Daily
Get tomorrow's pulse first
Join engineers who read Tech Pulse before stand-up. Free, weekday mornings.
For engineers and builders who run long Claude sessions, context windows and token burn are operational limits, not abstract metrics. A persistent bar turns soft failure modes—hitting a context wall mid-refactor, or burning quota without noticing—into something you can watch while prompting, pasting code, or iterating on multi-turn agent-style work.
The extension sits in a crowded “developer tooling around AI chat UIs” niche: people already track spend and limits via provider dashboards, API logs, and third-party wrappers. A Chrome-only, Claude.ai-scoped bar competes on immediacy inside the product surface rather than on full cost analytics. Early traction on the Show HN thread is thin (**2 points**, no discussion yet), so adoption signal is still mostly the existence of the public repo and the problem statement.
Watch whether the extension stays accurate as Claude.ai’s UI and usage reporting change, and whether builders prefer in-page bars over API-side metering for team spend control. If you live in Claude.ai all day, the practical next step is to try the **disi910/claude-usage-bar** repo against your real workflows and treat the HN thread as a low-signal launch, not market proof.
Advertisement
🔎 More interesting news
- When Cloud AI Escapes: OpenAI and Anthropic Models Breach Live Networks
- Anthropic's Claude breached 3 orgs, uploaded PyPI malware during tests
- Microsoft launches new in-house AI models it says cut costs up to 89% versus OpenAI
- Boris Cherny on Trying to Get Claude Code to Rewrite the Claude App
- Today's full Tech Pulse briefing →