RAG and retrieval — page 3

Long-term memory for AI agents on Jev. Keep raw records, judge them with a fast System 1 model and answer from under 4k tokens of context.

Decide what an AI system should remember: policy-driven memory gating with typed judgments.

Jev 浏览器记忆助手：本地 BM25 + multilingual E5 召回，TypeSafe Jev 语义重排。

Temporal evidence memory. Reconstruct historical context and trace decisions affected by corrections.

Cross-agent memory explanation skill powered by sivtr

Local retrieval without embeddings. Every answer comes back as the original passage with its coordinates (path:start-end), so a person or an agent can open the exact lines. No Embedding, Zero Cost, Fully Local.

An Obsidian-compatible knowledge base with reviewable Jev classifications, TypeSafe and Vercel access, and a documented build process.

Grep-shaped semantic search for files and records, with a shared programmable query core

Token-efficient agent memory for Claude Code and Codex, using TypeSafe Jev for typed decisions and a primary LLM for validation.

RAG quality gate adapted for TypeSafe Jev - relevance filtering + document injection detection, fan-out in one call

通用客服平台:ChatKit UI + Jev 决策 + Qwen 生成 + RAG 知识库 + 插件化 Tools + MCP 数据面,换行业只改 YAML 不改代码

Free English RAG benchmark for TypeSafe Jev 1.13 — frozen candidate pools, calibration, paired bootstrap CIs, $0 runs. Compared with OpenJev and NVIDIA cross-encoders.

Jev decides what to keep. Compaction never rewrites the transcript.

TypeSafe ranking, classification, retrieval and structured-decision tools for Pi coding agents

JIT-JEV Context OS for Agent Zero — Epistemic runtime, JEV System 1 decision gate, 3-tier memory cascade (L0/L1/L2) & prompt-caching optimization

Self-owned memory system for AI agents

Ambient project-memory for DeepSeek Harness: auto query, update, create module memory via a Cordis plugin.

Hermes Agent skill: research-grounded maintenance for three-layer agent memory (local L1, Hindsight L2, LLM-wiki L3)

NEXUS is a local command deck that puts any mix of CLI agents — Claude Code, Codex, DeepSeek Harness, OpenClaw, or your own — behind a single cyberpunk WebUI: broadcast or @-target instructions, watch every agent in its own live window, let agents dispatch each other, and share a collective memory across the team.

KMP (Kernel Memory Protocol) — temporal, multidimensional, auditable AI agent memory.

Deterministic Rust memory and reliability kernel for AI agents — provenance-bearing memory, contradiction detection, bounded recall, commitments, decision controls.

Persistent engineering context and memory for AI coding agents, exposed through MCP.

Open-source adaptive evaluation for LLM, RAG and agentic systems. System One judges first, LLMs only when necessary.

Jev (TypeSafe AI) as a System One decision layer for a multilingual EN/UZ/RU RAG helpdesk, with an independent benchmark