fast-jev-compaction
Claude Code plugin that replaces the compaction summary with pruning: Jev scores each old tool call, stale ones are cut and everything kept stays verbatim.
0 members list fast-jev-compaction, unchanged over the last 12 weeks, data as of 7 November 2026
Members who list fast-jev-compaction in their Stack, by week.
Members who list fast-jev-compaction
Be the first to add fast-jev-compaction
Nobody on Stackness runs fast-jev-compaction yet. Add it to your stack and you're the first member running it.
About fast-jev-compaction
fast-jev-compaction is a Claude Code plugin that modifies how conversation history gets compacted. Instead of summarizing old interactions, the plugin uses a scoring system called Jev to evaluate each tool call and its result. Calls identified as stale or outdated are removed or shortened, while everything the plugin decides to retain stays unchanged in its original form. The process happens in a single fast request.
The plugin is written in TypeScript and was developed by tamaratran. It is released under the MIT license, making it free to use and modify. The tool is designed for developers working with Claude Code who want more control over which parts of their conversation history get preserved during compaction, rather than having interactions automatically summarized away.
Checked 2 November 2026.
Nobody on Stackness runs fast-jev-compaction yet.
Sign up and add itPosts about fast-jev-compaction
- Can a decision model prune your coding agent's context? What fast-jev-compaction keeps, costs and drops
fast-jev-compaction prunes in a second, not 30. At its default threshold it keeps almost no tool results, and a scorer that always says no cuts as much.
- Coding agent harness, and which parts of it actually change the output
A harness is everything in a coding agent except the model. On strong models, two September 2026 studies found it moves the bill more than the pass rate.
- What is context compaction in coding agents, and what does it lose?
Compaction is the step where a coding agent summarises its own history to keep working: what it keeps, what it drops, and what OpenAI found in its summaries.