Claude Code / Codex / pi plugin that hands agent steps needing no text output to Jev (TypeSafe's judgment model) — measured p50 ~230 ms and ~$0.02 per 1,000 judgments, with typed escalation back to the LLM

TypeSafe Jev for DeepSeek Harness, the Model Context Protocol, and plain Node: typed judgments instead of prose, offline by default.

Automatic project memory for Claude Code: decisions, rules and dead ends saved to JEVMEM.md in your repo, brought back next session. In the Claude plugin directory and on the MCP Registry.

A calibrated context sieve for Claude Code: every tool result is judged by a System One model before it enters context.

Memory for AI agents that learns what is wrong and stops repeating it. Mark a memory wrong and it stops coming back; newer facts replace old ones. Local SQLite store and MCP server, persistent across sessions; hippo init wires it into Claude Code, Codex, Cursor, OpenClaw, OpenCode and Pi. Zero runtime deps, MIT, opt-in hosted TypeSafe Jev reranker.

Memory for AI agents that learns what is wrong and stops repeating it. Mark a memory wrong and it stops coming back; newer facts replace old ones. Local SQLite store and MCP server, persistent across sessions; hippo init wires it into Claude Code, Codex, Cursor, OpenClaw, OpenCode and Pi. Zero runtime deps, MIT, opt-in hosted TypeSafe Jev reranker.