smith-ctx-cursor

Summarize and selectively include files to manage Cursor context.

1|Updated Nov 25, 2025
One-click install
npx skills add https://github.com/tianjianjiang/smith --skill smith-ctx-cursor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: smith-ctx-cursor
Source: https://github.com/tianjianjiang/smith/tree/main/smith-ctx-cursor
Command: npx skills add https://github.com/tianjianjiang/smith --skill smith-ctx-cursor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cursor users struggle with context leakage in large sessions; this skill helps manage and prune context by summarizing and selectively including relevant files.

Core Features & Use Cases

  • Context-aware summarization with /summarize
  • @ mentions for including important files
  • @codebase discovery to locate relevant code
  • Use case: In a Cursor session with dozens of files, activate this skill to keep the AI focused on task-critical context.

Quick Start

Activate the skill in Cursor and trigger context optimization when your context grows beyond 60%.

Frequently Asked Questions about smith-ctx-cursor

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

FAQPage Schema
How do I manage Cursor context when it exceeds 60% in large workspaces?

To manage Cursor context exceeding 60%, apply a controlled summarization workflow using the /summarize prompt. This prunes context leakage in large workspaces by condensing active sessions and selectively restoring critical files.

What is the best way to keep an AI assistant focused on relevant code in a large session?

The best way to keep an AI assistant focused is by selectively including relevant files and summarizing inactive context. Using @codebase discovery guides context loading to locate relevant code references while pruning unnecessary background files.

How do I restore important files into the AI context after summarizing?

To restore important files into the AI context after summarizing, use @ mentions. This explicitly directs the context management workflow to reload specific task-critical files that were pruned during the summarization process.

Can I use @codebase discovery to locate relevant code for context loading?

Yes, you can use @codebase discovery to locate relevant code for context loading. This mechanism guides the AI to discover and reference specific code sections within large workspaces, ensuring focus and relevance matter.

Does context summarization work for large workspaces with dozens of open files?

Yes, context summarization works for large workspaces with dozens of open files. It applies specifically in Cursor sessions where context exceeds 60%, helping manage context leakage by focusing on task-critical references.