Fast Jev Compaction is a Claude Code plugin and npm library that replaces traditional conversation summarization with selective context pruning. Instead of rewriting older messages, it asks Jev to decide which tool calls and results are still useful. Stale tool interactions can be removed or truncated, while retained content stays verbatim and in its original order. User and assistant text is preserved rather than compressed into a generated summary. Recent messages can be pinned so they are never discarded during compaction. The system progressively reduces oversized context before scoring it and falls back to Claude Code’s built-in compaction when necessary. It is designed to preserve exact paths, errors, commands, and constraints that ordinary summaries might lose.

Features

  • Jev-based context relevance scoring
  • Verbatim message preservation
  • Tool call and result pruning
  • Configurable recent-message protection
  • Progressive context-size reduction
  • Claude Code plugin and npm library

Project Samples

Project Activity

See All Activity >

License

MIT License

Follow Fast Jev Compaction

Fast Jev Compaction Web Site

Other Useful Business Software
Train ML Models With SQL You Already Know Icon
Train ML Models With SQL You Already Know

BigQuery automates data prep, analysis, and predictions with built-in AI assistance.

Build and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
Start Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of Fast Jev Compaction!

Additional Project Details

Programming Language

TypeScript

Related Categories

TypeScript Artificial Intelligence Software

Registered

23 hours ago