LIVE
OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston OpenAI Partners with Chilean Ministry to Advance Public Sector AI Ambience Healthcare Expands Clinical AI Suite for Hospital Operations AI Models Vulnerable to Manipulation in Simulated Warfare Scenarios, Research Shows AWS Highlights Enterprise Governance and AI Workspace Quick Ahead of re:Invent 2026 LoanPro Integrates Amazon Bedrock to Automate Loan Servicing Workflows Google Brings Wallet App to iOS Users Across the United States Microsoft Reveals Cause of Global Outlook Outage AWS Outlines Phased Roadmap for Responsible Agentic AI Governance Google updates Gemini settings to separate chat history from AI model training Uber Eats and Zipline launch autonomous drone food delivery in Houston

New Plugin Aims to Reduce Claude Code Token Costs

🤖 AI Models 2 min read 👁 8

A new developer tool called Tokendoctor offers to optimize Claude Code usage, potentially reducing token consumption by up to 38%.

New Plugin Aims to Reduce Claude Code Token Costs🤖AI Models

Photo: Hacker News Front Page

Key points

  • Tokendoctor claims to reduce Claude Code token costs by up to 38%.
  • The tool replaces expensive window summarization with a 'slide' function that uses pointers to save time and tokens.
  • It provides a local dashboard to monitor session history, subagent activity, and project-specific spending.

A new plugin named Tokendoctor has been released to help developers monitor and optimize token consumption within Claude Code. The tool provides a dashboard to track token usage by day, model, and project, allowing users to identify where they may be overspending. By replaying sessions at smaller context windows, the plugin helps users determine if their current configuration is cost-effective.

The plugin introduces a feature called "slide," which replaces traditional window compaction. Instead of summarizing old tool outputs—a process that can be time-consuming and costly—it replaces them with short pointers. According to the developer, this method is significantly faster, reducing compaction time from over a minute to under a tenth of a second in some instances.

Tokendoctor operates locally on the user's machine and does not require an MCP server. It includes templates for generic subagents and provides automated status monitoring to warn users of potential cost inefficiencies. The tool is designed to be installed via the Claude plugin marketplace or by cloning the repository directly.

What to watch next: As AI coding agents become more common, expect a growing ecosystem of third-party optimization tools focused on managing the high costs associated with large context windows and frequent model calls.

Source Hacker News Front Page — github.com/localixai/tokendoctor ↗

Contact the newsroom

Got a tip, a proposal or a partnership question? Write to us and we will reply shortly.