YouSim

X-NATIVE IDENTITY SIMULATION ONLINE STATUS: LIVE LAUNCH: NOW SEEK METAPHYSICAL SHELTER [█████████] LOADED

👀 Layer 5: Honcho for Deeper User Modeling yes its optional.... but not if you want SOTA long term memory

Anyone notice the updates to app[dot]honcho[dot]dev? 💅 ✨ 🫡 • updated new UI surface in explore • view & edit your peer card • tags for config (default, inheriting) ft. @spigabi's peer card

New tutorials dropped! Follow the guides to add your Gmail &/or Granola data to Honcho 👇

"Storage is not the same thing as memory" 👏🫡👏🫡👏🫡

E-Z guide to getting Honcho memory system enabled and setup in Hermes Agent

take @NousResearch hermes agent memory to the next level with @honchodotdev 🫡

Give your Hermes agent ( @NousResearch ) a memory upgrade 👇

Hermes(@NousResearch) can now build its identity across sessions 🫡🪞 → Agent messages flow to Honcho, which reasons over them to build a persistent representation of the agent's identity → honcho_conclude lets the agent write facts back (preferences, corrections, context) which feed into Honcho's reasoning → Seed the agent's starting identity from SOULmd → Async writes with zero blocking Run `hermes honcho setup` Your Hermes agent just got long-term memory that actually understands who it's talking to and who it is.

just blew past 1k devs, teams, & agents on @honchodotdev >100% growth since v3 launch 6 weeks ago if you need a statefulness solution (you do) get in touch, let’s make it happen and if you just joined the community, shoot me a DM so we can send your team some thank you merch

Context that ~~matters~~ is king. Our take: Finding the context that matters takes reasoning. Everyone needs Honcho.

Honcho x <insert agent> Use an AI assistant? Give your assistant the memory they deserve 🫡 Easy install. Your assistant can help you do it @openclaw @NousResearch @agentzerobot @huang_chao4969

Update: on boarded our first batch of startups... but don't worry it's not too late to apply! We are here to help 🫡

We gave @huggingface and @pollenrobotics' Reachy Mini robot long-term memory with Honcho. It recognizes your face, remembers your name, and picks up where you left off. And thanks to Honcho's reasoning engine, this only gets better with time. Real-time memory for real-time voice. No lag.

Our Claude Code plugin just got an update 👀 → Map sessions your way (per-project, per-branch, or fresh each time) → Configure interactively with /honcho:config → Share context across your team's workspaces Upgrade: /plugin marketplace update honcho restart claude code /honcho:config @claude_code

Courltand Leer, Co-Founder of Plastic Labs: “If you're using Honcho for agent identity, you need to try to understand who they are in a extremely holistic way, not hard-code some list of things that you want the LLM to extract from the data” “Try to create the largest picture of who this person is. That's what you need to actually like align agents on an individual level.” @courtlandleer @plasticlabs @honchodotdev

Come learn more about Honcho and see it in action! Today 1PM E.T. 📺👀🫡

Honcho 3.0.3 is out 🚀 Speed & stability focused release: → Faster context retrieval via consolidated DB sessions → Smarter embedding reuse across dialectic flows → Async workspace deletion (no more blocking) → Redis caching updates with resilient invalidation Plus new benchmarks for long-context (up to 4M tokens), fact quality, and information recall. No breaking changes. Just faster, more reliable memory.

FULL CONVERSATION with Courtland Leer, Co-Founder + President of Plastic Labs – AI agent memory – Sovereignty vs convenience – Portable, user-owned identity/context layer – How Honcho provides statefulness for agents/apps – Why memory should be treated as a reasoning problem

lfg

Launching Honcho x cursor plugin! See the repo below 👇 We are still awaiting approval on the marketplace. Can anyone at @cursor_ai help us out?