You already own the agent. The only reason setup has ever meant a wall of commands is that no tool has let you just ask for it. Tell your agent "install brethof-brain," and it does the whole job — fetches the client, wires it into your tools, checks its own work. No server to run, no database to stand up, no config files to edit. One plugin, a few hundred lines long, short enough to read end to end if you're the kind who verifies what leaves your machine.
1. Sign up at brethof.ai/account, copy your API key. The one human step, and it's the same step every application on earth asks for. No credit card on the free plan.
2. Say "install brethof-brain." That's the whole sentence — the one step you don't have to read about.
3. Verify, if you want to. Ask your agent to search memory for anything, even something trivial. A well-formed answer — even "no results" — means it's connected. That's the entire test.
Get your API key Full setup guide
Verify, if you want to. Ask your agent to search memory for anything, even something trivial. A well-formed answer — even “no results” — means it's connected. That's the entire test.
The memory doesn't live in files you have to maintain. It's two stores that work as one: a curated record of the current truth, dense and up to date — and the complete raw transcript of every session, kept whole and searchable by exact words, by meaning, or through the graph of files, services, decisions and errors your work touched. All three folded into one answer. Recall never depends on what survived a summary or fit in a context window. And the memory builds itself. After each exchange, a curator reads the turn and rewrites what's known: decisions saved, changed facts folded in, claims reality has overtaken deleted. Seconds, not sessions. Nothing has to be filed by hand. When you do want something kept on the spot, you say so, and it's saved. Once a week, a healer reads the entire memory at once and does what no single exchange can: merges records that forked apart, absorbs duplicates written months apart, and reconciles the places where memory disagrees with itself.
Using the GLM coding plan? That runs inside Claude Code — full memory loop, same plugin.
And the things you killed stay dead. Reversed decisions sit in the graph flagged as dead ends, and an agent steering toward one is stopped mid-prompt. Your memory is stored, indexed and searched in the EU — the one exception is a curation pass to a provider whose terms say inputs are transient and never trained on. What leaves your machine: your messages, your agent's replies, its thinking, and one-line markers saying which tool it used. What never leaves: the contents of files it read and the output of commands it ran.
One memory, every agent you run — several at once, or a whole team on the same project. Projects are just names: say one when saving or searching, and it exists from that moment. Rules you set, for a project or for everything, shape how every agent works and load automatically into every session.
Local speech-to-text that learns your voice. Perpetual licence. Our flagship.
PAID · flagship
Long-term memory for your AI agents — full-text + vector + graph. Briefed at session start, recalled on every prompt, archived automatically.
PAID · free tier
Print-ready digital models. GLB and OBJ included. Lifetime access.
PAID · digital catalog
Five printers and real capacity. No self-serve shop yet — tell us what you need and we quote it by email.
BY ARRANGEMENT · email us
Our YouTube channel. A cyber-tiger host walks through local AI tools and what they actually do.
CHANNEL · live
Curated GitHub lists for AI coding agents, MCP servers, local AI and Linux for AI. Every entry carries a source link.
FREE · curated
Long-form how-tos for local AI on Linux, Windows and macOS, with the configuration files included.
FREE · live
Negative-curation: practices and tools that waste your time, ranked. Receipts required.
FREE · live
Who we are, why we build privacy-first AI, and what we won't do.