What did Hlido score ClaudioDrews/memory-os?

ClaudioDrews/memory-os scored 73/100 (STEADY) in Hlido's independent, hands-on review.

Is ClaudioDrews/memory-os reliable?

ClaudioDrews/memory-os is rated STEADY (73/100) by Hlido. ClaudioDrews/memory-os — A 7-layer memory operating system for Hermes Agent — persistent memory with Qdrant, structured facts, fabric recall, auto-curated wiki, and surgical context injection. Runs locally, any LLM provider.. Repo-surface review: measured from the public reposito

How much does ClaudioDrews/memory-os cost?

Hlido recorded the pricing model as: Open source · Subscription. Derived from evidence Hlido holds, not vendor-supplied; verify current prices on the vendor's pricing page.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

ClaudioDrews/memory-os

Name: ClaudioDrews/memory-os review
Item: ClaudioDrews/memory-os
Rating: 73
Author: Hlido Editor

Infrastructure · tested 2026-08-27 · by the Hlido desk, not the vendor

In short: A serious, MIT-licensed local-first memory layer for agents with real traction (1.3k stars) — but no tagged release and thin surfaced install signals temper the maturity read.

7 PASS · 3 FAIL of 10 public-surface claims

Visit ClaudioDrews/memory-os → scorecard.json

Quick answer

ClaudioDrews/memory-os scores 73/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-08-27). STEADY (73, repo-surface ceiling). Pricing: Open source · Subscription (free entry point documented).

Hlido Editor · 2026-08-27

Memory OS is an ambitious, genuinely useful-sounding piece of agent infrastructure: a seven-layer persistent-memory stack (Qdrant vectors, structured facts with trust scoring, a self-curating wiki, semantic recall) that runs entirely on the user's machine and stays provider-agnostic across OpenRouter, OpenAI, Anthropic and Ollama. The repo signals back the pitch — 1,272 stars, MIT license, Python, CI present, and a commit within the last ~36 days. Two things keep it out of the top band on the repo surface: there are no tagged releases (so "stable" is a judgment call, not a versioned promise), and our checklist did not detect standard install/usage markers even though the README advertises a one-command curl installer — a documentation-shape gap worth closing. It's built to embed into one specific host (Hermes Agent) via a plugin, so treat it as a memory component to wire in, not a drop-in for arbitrary stacks. Repo-surface read only; we did not run the memory pipeline hands-on.

Why STEADY

STEADY (73, repo-surface ceiling). MIT licensing, Python with CI, strong traction (1,272 stars) and recent activity clear STEADY. VITAL is withheld because repo signals cannot verify the memory pipeline works as claimed, and both the absence of tagged releases and the missing install/usage markers hold it below the top of the band.

Public-surface checklist

PASS Repo reachable (required) — GH API 200 for ClaudioDrews/memory-os
PASS Readme present (required) — README length 10223
PASS License present (required) — MIT
FAIL Install documented (required) — no install/usage markers
PASS Active 12mo (required) — last push 36d ago
FAIL Releases present — no releases
PASS Community traction — 1272 stars
PASS Ci or tests — 1 workflow file(s)
PASS Recent commit 90d — last push 36d ago
FAIL Agent consumable — none found

What we saw

1 screenshot captured by the Hlido engine during the reviewed run (run-3e75fd0ccfef6c0a-github-com). Our own captures — not vendor marketing material.

ClaudioDrews/memory-os — run screenshot 1 (home.png) — home.png

What it does well

Local-first, self-hosted — memory infrastructure runs on the user's own machine
Provider-agnostic across OpenRouter, OpenAI, Anthropic and Ollama (no vendor lock-in)
Strong community traction for an infra project (1,272 stars, MIT)
CI present and actively maintained (last push within ~36 days)

What it fails at

No tagged releases — no versioned stability contract
Install/usage markers not detected on the repo surface despite an advertised one-command installer
Coupled to Hermes Agent as its host rather than exposing a general agent interface
Functional quality of the memory pipeline unverified from the repo surface

Best for

Developers building on Hermes Agent who want persistent, local memory
Privacy-conscious setups that need on-device memory with no memory subscription
Teams wanting provider-agnostic memory they can self-host

Not recommended for

Teams needing a versioned, release-pinned dependency
Non-Hermes stacks wanting a plug-and-play memory API
Anyone who needs verified behavioral guarantees before adopting

Pricing & access

ModelOpen source · Subscription
Free entry pointYes — a free tier or open-source edition is documented

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-07-16.

Compared to

Letta local-first self-hosting
Letta (formerly MemGPT) is the better-known, more general stateful-agent framework with a hosted path; Memory OS is a self-hosted, local-first memory layer aimed at one host (Hermes Agent). Choose Letta for a general, managed statefulness layer; choose Memory OS for on-device, provider-agnostic memory you control end to end.

Agent relevance

CLI

Agentic-Commerce Readiness 27/100 · SURFACE-ONLY

Independent readiness for agent delegation & transaction. How it’s scored · check live

Embeds into Hermes Agent as a memory plugin; installed and run locally via a setup script rather than consumed as a standalone programmatic API. No hands-on behavioral test performed.

Agent-friendly score: 5/10

Evidence

— source
— source

scorecard.json · transparency passport · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: repo-surface-track+editorial-narrative-v2 · Methodology version 2026.05 ·

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/claudiodrews-memory-os.svg)](https://hlido.eu/check/?agent=claudiodrews-memory-os)

HTML

<a href="https://hlido.eu/check/?agent=claudiodrews-memory-os"><img src="https://hlido.eu/badge/claudiodrews-memory-os.svg" alt="Hlido trust score"></a>