skills.rest
sigridjineth
sigridjinethCOMMUNITY
Explore all skills by this author →
GitHub

cdp-context-editing

Summarize and extract facts from long conversation contexts to reduce token usage.

5|Updated Dec 14, 2025
Search Tags:#summarization#context management#token optimization#cost savings#Claude API#fact extraction#long conversations
One-click install
npx skills add https://github.com/sigridjineth/anthropic-assignment --skill cdp-context-editing
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cdp-context-editing
Source: https://github.com/sigridjineth/anthropic-assignment/tree/main/skills/cdp_context_editing
Command: npx skills add https://github.com/sigridjineth/anthropic-assignment --skill cdp-context-editing

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the critical challenges of Claude forgetting earlier messages in long conversations, high token costs, and the complexity of managing the context window. It helps users maintain conversational coherence and reduce operational expenses.

Core Features & Use Cases

  • Summarization: Compress older turns while preserving meaning, ensuring Claude stays focused.
  • Fact Extraction: Pull out key facts to persistent storage, preventing important details from being lost.
  • Selective Retention: Keep important messages verbatim, ensuring critical information is always present.
  • Use Case: For a 20-turn customer support conversation, this Skill can reduce token usage by 60-70% while ensuring Claude "remembers" the customer's initial problem and preferences.

Quick Start

Explain how Context Editing can help manage a 20-turn conversation about financial planning without losing key details or incurring excessive token costs.

Frequently Asked Questions about cdp-context-editing

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

FAQPage Schema
How do I reduce token costs in long Claude API conversations?▼

Token cost in long conversations grows with each turn. Context editing compresses older messages through summarization and fact extraction, reducing token usage by 60-70% while preserving key details Claude needs to stay coherent.

Can I keep Claude from forgetting earlier messages in a 20+ turn conversation?▼

Claude's context window limits what it can see. Selective retention stores critical messages verbatim and extracts persistent facts, ensuring Claude remembers your initial problem and preferences across many turns without losing context.

What's the best way to manage context in token-cost constrained deployments?▼

Context editing optimizes token usage by dynamically managing which messages stay verbatim, which compress into summaries, and which extract into persistent facts. This approach balances cost, coherence, and memory across long deployments.

How does fact extraction prevent important details from being lost in conversations?▼

Fact extraction pulls key information from earlier turns into structured storage outside the conversation flow. These persistent facts remain available to Claude even when older messages compress or leave the context window.

Do I need to manually edit context, or can this process be automated?▼

This Skill automates the editing process—it identifies which turns to compress, which facts to extract, and which messages to retain based on conversation length and token constraints, removing manual intervention.

Related Skills

Explore More Software Engineering Skills→
arturogj92arturogj92

Engagement Learning

Log outreach interactions and extract response patterns from Markdown-based workflows.

Community
Basic
ChakshuGautamChakshuGautam

memory-systems

Stores and retrieves agent memories via vector store and temporal knowledge graph with entity/time filters.

Community
Advanced
kivo360kivo360
1

moai-cc-memory-arch

Manage Claude Code session memory with just-in-time retrieval and context summaries.

Community
Advanced
Prathmesh2000Prathmesh2000
1

token-analyzer

Analyze and estimate token usage for LLM API calls.

Community
Intermediate