When should I use /clear instead of /compact?

Use /clear when switching to an entirely different project or unrelated topic to start with zero tokens. Use /compact when you are continuing the same task but need to free up window space.

How does the /compact command help in Claude Code?

The /compact command summarizes your conversation history to preserve key decisions and file awareness while reducing the total token count by 50-80%, leading to faster AI response times.

What are the early signs of context pressure?

Signs include noticeably slower response times, Claude forgetting earlier constraints, repeated questions about previously read files, and the token usage bar exceeding 70%.

How do subagents assist with context management?

Subagents run in their own separate context windows. Delegating verbose tasks like file exploration or test execution to a subagent prevents your main session from becoming cluttered with unnecessary output.

Context Optimization

Name: Context Optimization
Author: hgeldenhuys

byhgeldenhuys

•

生产力与工作流

Optimizes Claude's context window usage through strategic history compaction, token budgeting, and efficient session management.

关于

Context Optimization is a specialized skill designed to help developers maintain high performance and lower costs during extended Claude Code sessions. It provides actionable guidance on managing the 200K token context window using commands like /compact and /clear, identifying performance bottlenecks, and strategically delegating tasks to subagents. By following the established workflows for context budgeting and session organization, users can ensure Claude retains critical project information without suffering from the latency or forgetfulness associated with context saturation.

主要功能

Subagent delegation strategies to isolate verbose tasks
Visual and behavioral context pressure indicator checklists
Standardized session management and multi-session workflow patterns
1 GitHub stars
Intelligent context compaction to preserve key decisions while reducing tokens
Real-time usage monitoring with /cost and token estimation guides

使用场景

Maintaining AI performance during large-scale codebase refactoring and migrations
Budgeting context for complex feature implementations that span multiple files
Optimizing long-running research or documentation tasks through strategic subagent use

关于

主要功能

Subagent delegation strategies to isolate verbose tasks
Visual and behavioral context pressure indicator checklists
Standardized session management and multi-session workflow patterns
1 GitHub stars
Intelligent context compaction to preserve key decisions while reducing tokens
Real-time usage monitoring with /cost and token estimation guides

使用场景

Maintaining AI performance during large-scale codebase refactoring and migrations
Budgeting context for complex feature implementations that span multiple files
Optimizing long-running research or documentation tasks through strategic subagent use