skills.rest
northseadl
northseadlCOMMUNITY
Explore all skills by this author →
GitHub

context-compression

Compress LLM context windows using anchored iterative summarization and probe-based evaluation.

Updated Feb 3, 2026
Search Tags:#optimization#llm#compression#token usage#context window#agent performance
One-click install
npx skills add https://github.com/northseadl/skillwisp --skill context-compression-northseadl
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-compression
Source: https://github.com/northseadl/skillwisp/tree/main/skills/%40muratcankoylan/context-compression
Command: npx skills add https://github.com/northseadl/skillwisp --skill context-compression-northseadl

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill addresses the challenge of managing large conversation histories and codebases that exceed LLM context window limits, preventing information loss and improving agent performance.

Core Features & Use Cases

  • Context Compression Strategies: Implements multiple techniques like Anchored Iterative Summarization and Opaque Compression to reduce token usage.
  • Tokens-Per-Task Optimization: Focuses on minimizing total tokens consumed per task, not just per request, by preserving critical information.
  • Use Case: When an AI agent is working on a large codebase or a long-running session, this skill ensures it doesn't "forget" crucial details by intelligently summarizing and compressing the conversation history.

Quick Start

Use the context-compression skill to summarize the last 5000 tokens of the current conversation.

Frequently Asked Questions about context-compression

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

FAQPage Schema
How do I reduce LLM context window usage for long-running agent sessions?▼

Tokens-per-task optimization minimizes total tokens consumed per task by preserving critical information through compression strategies, ensuring agents maintain performance without forgetting crucial details over long sessions.

How do I compress conversation history in large codebases without losing information?▼

Probe-based evaluation tests compressed context to verify that critical information is retained, preventing context degradation and ensuring the agent maintains accurate performance on large codebases.

What is the best way to optimize token usage for AI agents working on large codebases?▼

The best way to optimize token usage is prioritizing tokens-per-task over tokens-per-request, using compression strategies like anchored iterative summarization to manage large codebases efficiently within context window limits.

Why does my AI agent forget crucial details when processing large conversation histories?▼

You can summarize specific token ranges of a conversation by using the context-compression skill, such as summarizing the last 5000 tokens to manage and reduce current context window usage effectively.

Related Skills

Explore More Software Engineering Skills→
bthillerupbthillerup
1

context-compression

Compress large conversation histories and codebases exceeding context window limits.

Community
Advanced
jax2730jax2730

context-compression

Compress extensive conversation histories for AI agents using iterative summarization.

Community
Advanced
ChakshuGautamChakshuGautam

context-compression

Compress agent session context using anchored iterative summarization and probe-based evaluation.

Community
Advanced
wangyouanwangyouan

context-compression

Compress large token contexts to minimize information loss in agent sessions.

Community
Advanced