post-compact

Recover session context after compaction using local history search.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Roger-Parkinson-EHP/dev-experience --skill post-compact
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: post-compact
Source: https://github.com/Roger-Parkinson-EHP/dev-experience/tree/main/claude-code/skills/post-compact
Command: npx skills add https://github.com/Roger-Parkinson-EHP/dev-experience --skill post-compact

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill ensures you don't respond to users until the full context has been recovered after a compaction, preventing confusion and misalignment.

Core Features & Use Cases

  • Context Recovery: Identify current tasks, recent changes, open issues, and user preferences.
  • Pre-Response Validation: Confirm understanding before replying to the user.
  • Use Case: After a compaction summary mentions a specific topic, recover the actual context by inspecting recent exchanges to inform the next steps.

Quick Start

cd ~/.claude/services python conversation-search.py "INSERT_TOPIC_FROM_SUMMARY" --max 5 --verbose

Frequently Asked Questions about post-compact

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

FAQPage Schema
How do I recover context after a conversation compaction?

Context recovery after compaction restores your active tasks, recent changes, and decisions by inspecting the compaction summary and searching recent exchanges. Use conversation-search to retrieve relevant context before responding to the user, ensuring continuity and alignment.

What happens to my session context during a compaction event?

During compaction, your conversation history is summarized to free up space. A compaction summary captures key points, but full context—including specific file changes, open issues, and user preferences—must be recovered using local history search to continue effectively.

When should I apply post-compact context recovery?

Apply context recovery immediately when a compaction summary appears or when a session starts with a prior summary. Recovering context before your next response prevents misalignment, ensures you understand current tasks, and restores awareness of recent decisions and modifications.

How do I use conversation-search to retrieve lost context?

Run conversation-search with a topic extracted from the compaction summary to extract relevant exchanges. The tool inspects local history, identifies current task state and recent changes, and delivers recovered context ready for your next response.

Can I restore specific file modifications and errors from a compaction?

Yes. Context recovery extracts modified files, error states, and technical decisions from recent exchanges after compaction. Conversation-search retrieves these details so you can resume work without losing track of what changed or what failed.