Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Public story · 2026-03-02 · source-backed

Agent Memory Becomes Its Own Product Category

Issue

Confidence

source-backed

Sources

4

Redaction

redacted

Story

Four new frameworks in one day — each with 6K-12K stars — signal that agent memory has graduated from "feature" to "product category":

memU (12.3K stars) — Memory for always-on proactive agents with filesystem-like hierarchy and proactive intent prediction. PostgreSQL + pgvector backend. GitHub
Memori (12.3K stars) — SQL-native memory layer with three hierarchical levels (Entity/Process/Session). Zero-config, framework-agnostic, zero-latency background augmentation. GitHub
MemOS (6.1K stars) — Memory Operating System with graph-structured inspectable memory (Neo4j + Qdrant). 72% lower token usage claim. "Skill memory" for cross-task reuse. GitHub
Zep Graphiti — Temporal knowledge graphs outperforming MemGPT on Deep Memory Retrieval (94.8% vs 93.4%), 90% latency reduction. Accepts JSON business data, not just chat. arXiv

Related stories

Each link below shares sources, entities, or timing with this story.

Temporal partners with OpenAI / Shared entities / Same source domain / Shared topic / What happened next
Agent Infrastructure Fills Five Of Product Hunt's Top Seven
Linked by a graph relationship (Temporal partners with OpenAI); both cover GitHub, Memory, MemOS, MemTensor; reported by the same outlet (github.com).
Shared entities / Same source
Projects & Repos
Both cover Memori, MemoriLabs, MemOS, MemTensor; cite the same source (GitHub, GitHub, GitHub).
Shared entities / Same source / Shared topic / What happened next / Tension
Memory Wars
Both cover GitHub, MemOS, MemTensor; cite the same source (GitHub); overlapping topics (agent, claim, memory, star).
Memori benchmarked against mem0 / Shared entities / Same source / Shared topic / What happened next / Tension
Memori v3.3.3: Agent-Native Memory Layer. 81.95% Accuracy at 4.97% Context. 14.3K Stars.
Linked by a graph relationship (Memori benchmarked against mem0); both cover Memori, MemoriLabs; cite the same source (GitHub).
MemOS released OpenClaw / Shared entities / Same source domain / Shared topic / What happened next
claude-mem crossed 80K stars. Persistent memory across 28+ agent runtimes from one install
Linked by a graph relationship (MemOS released OpenClaw); both cover GitHub, Memory; reported by the same outlet (arxiv.org, github.com).
Desktop "cowork" shells are turning coding CLIs into interchangeable backends.
Linked by a graph relationship (MemOS released OpenClaw); both cover Session, SKILL; reported by the same outlet (github.com).
Memori benchmarked against mem0 / Shared entities / Same source domain / Shared topic / What happened next / Tension
Two agent-memory projects publish directly conflicting benchmark claims on different splits.
Linked by a graph relationship (Memori benchmarked against mem0); both cover GitHub, MemGPT; reported by the same outlet (github.com).
Temporal partners with OpenAI / Shared entities / Same source domain / Shared topic / What happened next
Garry Tan's GStack Hits 10K Stars in 48 Hours: The Thesis That Structured Prompts Are All You Need
Linked by a graph relationship (Temporal partners with OpenAI); both cover GitHub, SKILL; reported by the same outlet (github.com).

Source trail

GitHub GitHub GitHub arXiv

Entities

Accepts JSON Deep Memory Retrieval Entity GitHub MemGPT Memori MemoriLabs Memory Memory Operating System MemOS MemTensor Neo4j NevaMind PostgreSQL Process Qdrant

Provenance

Canonical issue: Ramsay Research Agent — 2026-03-02
AI generated: no
Story unit: 2026-03-02-agent-memory-becomes-its-own-product-category
Labels: source-backed, canonical briefing excerpt