context-manager

Analyze token usage and guide progressive file loading for Claude Code.

1|Updated May 5, 2025
One-click install
npx skills add https://github.com/jdeweedata/circletel --skill context-manager-jdeweedata
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: context-manager
Source: https://github.com/jdeweedata/circletel/tree/main/.claude/skills/context-manager
Command: npx skills add https://github.com/jdeweedata/circletel --skill context-manager-jdeweedata

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill helps software teams manage Claude Code's token budget by analyzing token usage and guiding selective file loading. It reduces context overflow risks in large codebases.

Core Features & Use Cases

  • Token usage analysis to quantify how much content is loaded into context.
  • Progressive loading guidance using line ranges and modular file loading.
  • Easy integration with .claudeignore and optional .context-notes to tailor project-specific loading strategies.
  • Real-world workflows for large-scale refactors, debugging, and feature work.

Quick Start

Install and run the included analysis script in your project:

  • Run: python ~/.claude/skills/context-manager/scripts/analyze_tokens.py .
  • Create or customize a .claudeignore via the provided template in assets/.
  • Begin using targeted queries like "Show me the structure of src/" and "Load src/api.py lines 100-150" to practice progressive loading.

Frequently Asked Questions about context-manager

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I manage Claude Code token budget for large codebases?

You manage the Claude Code token budget by running an included Python script to analyze token usage, then applying progressive loading with line ranges to prevent context overflow in large codebases.

What is progressive loading and how does it optimize context for multi-file tasks?

Progressive loading optimizes context by guiding selective file loading using targeted line ranges, which reduces context overflow risks and conserves your token budget during multi-file tasks.

How do I analyze token usage in my Python project before loading files?

You analyze token usage by running the analyze_tokens.py script in your project directory, which quantifies how much content will be loaded into Claude Code's context.

Does Claude Code work with .claudeignore to manage token budgets effectively?

Yes, Claude Code works with .claudeignore files to manage token budgets by excluding unnecessary files from context loading, which can be customized using a provided template.

Can I load specific line ranges of a file to reduce context overflow in Claude Code?

Yes, you can load specific line ranges of a file to reduce context overflow in Claude Code by using targeted queries like asking to load a file's specific lines.

Do I need any special setup to start analyzing tokens for my development workflow?

No special setup is needed to start analyzing tokens for your development workflow beyond running the included Python script, though you can optionally add a .context-notes file.