Repo · Context & memory

Fast Jev Compaction

Decide which tool interactions to retain.

Open github.com ↗

From the repository

Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim.

How builders describe it

Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim.

Prune old tool activity from an agent conversation while preserving ordinary message text and the tool evidence selected for retention.

Claude Code compaction that deletes instead of summarising. Alex Volkov: a Claude session went from nearly 1M tokens to 86K in about one second.

This uses @typesafeai Jev model as a plugin in Claude to review all the unnecessary tool calls, and it takes 1s to run! Literally 1 second to take a Claude session from nearly 1M to 86K tokens. Install prompt: Install, and configure: — @altryne (7.2K likes)

The decision Jev makes

Keep or discard a tool result, or answer whether a stopping condition is met.

Where it fits

Agents drown in their own context. These projects use Jev to decide what to keep, what to drop and when a loop is actually finished. All 66 context & memory projects →

Related projects