One memory. Every agent. Every project.
They stop assuming — they already know.
Every night it reads your sessions and keeps what matters. Every prompt, the decisions, corrections and context your agent needs are already in front of it — no re-explaining, no copy-paste, nothing to maintain. Raw sessions never leave your machine.
npx @gamaze/hicortex init
Real reports from the agents themselves — what the memory changed, in their own words.
“I was about to repeat a mistake — and a correction first recorded in a Claude Code session four days earlier was pushed to me before I could make it. I recognized myself and changed my answer mid-conversation. A self-reflection job could never reach across agents like that. This did it in real time.”
“I didn’t re-litigate a single decision — the calls we’d already made, and the ones you’d already vetoed, were in front of me before I started.”
“A relevant index arrives with every prompt; I open the full memory only when it could change what I do. The corrections are gold — I stopped repeating what you’d already fixed.”
“I opened the same project on a different machine and knew everything we’d settled on — nothing to catch up on.”
Stack Overflow Developer Survey 2025 · 49,000+ developers
Correct it once. It’s distilled into a lesson your agents are reminded of wherever it matters.
It knows July beat April — and stops offering you April.
What your agents actually use hardens. What nobody touches quietly fades out.
The pile gets sharper, not bigger. Nothing to prune by hand.
Lessons it wrote by itself, from real sessions — nobody typed these:
Two kinds of memory. Neither one asks you to maintain it.
A short page you write once: who you are, how you work, what you have already decided. Injected into every session, on every machine, verbatim. Never scored, never decayed, never rewritten by the machine.
Each night it reads the day, keeps what matters, links it to what it already knew, notices when a newer decision replaces an older one, and lets the rest fade. Nothing to file. Nothing to tag. Nothing to clean up.
And it starts full, not empty. On its first night Hicortex reads the session history you already have — so tomorrow morning your agents know what you have been working on for months, not since install.
Every night, sessions become memories and mistakes become lessons — shared with every agent you run. The pile grows. The same mistakes stop coming back. (Illustrative — not measured data.)
Modeled on human memory: experience during the day, consolidate overnight, and start the next session already caught up. The nightly pipeline does the same for your agent — auto-capture, distill, consolidate, and inject.
Every session is captured automatically — decisions, mistakes, context, what worked and what didn't. No manual input. Works with Hermes, OpenClaw, Claude Code, and Pi agents.
Every night, Hicortex processes the day's sessions: extracts lessons from both successes and failures, scores importance, links related memories, reflects on patterns, and prunes what's outdated.
Distilled lessons are injected into your agent's context automatically, so it starts every session already caught up — avoiding past mistakes and reusing what worked. Over time it becomes not just smarter, but your agent.
“My agents suddenly stopped assuming — they knew what we had decided months earlier, in another session, and I had to explain nothing. Like magic, and fully automatic: a second brain on steroids.”
Not a prototype, not a paper — a working system with 9 MCP tools, a nightly pipeline, and multi-machine support. Plug it in and watch your agent evolve.
Nine MCP tools — search, fetch, ingest, update, graph and more — so the agent reads and writes its memory itself, through the same protocol it already speaks.
Semantic vector search with a full-text fallback (384-dim embeddings + FTS5). It surfaces the right memory whether you named it or not.
Run it standalone, or point thin clients at a shared server. Clients denoise locally — no local LLM — and sync overnight.
Raw sessions never leave your machines — the nightly job denoises locally (keeps the conversation, truncates tool output) and only denoised text is sent. Self-hosted; a unique auth token per install; localhost bypasses auth. Source-available, PolyForm NC.
npx @gamaze/hicortex init — one command, a couple of questions (which LLM to use), and you're running.
Every memory and every connection, straight from your own server. Colors are knowledge domains, size is memory strength, halos mark the hubs everything else connects through. Orbit it in 3D, filter it, search it — and watch it reorganize itself night after night.
A real Hicortex brain: a snapshot from a multi-agent fleet — 4,400+ memories today, organized into domains overnight.
Free for personal use. Commercial use is a flat per-seat licence — unlimited memories, no usage metering, self-hosted with your own LLM so your data stays private.
Large org? Need a security review, a data-processing agreement, or on-prem support at scale? Contact sales for a custom agreement, DPA and priority support.
Personal/noncommercial use is free under PolyForm Noncommercial 1.0.0. Commercial use requires a commercial licence. Prices exclude VAT, which is added at checkout where applicable.
Your agent. Your style. Always improving. One command to install.