🤫husshhussh
🤫husshhusshOnePuppy
The hussh blog · Open problems

The hard parts, explained plainly.

Original hands-on essays about the engineering questions behind 🤫 Agent One. No borrowed voice, no hidden roadmap claim, and no pressure to agree before you inspect the work.

Research notesAll research
Six essays

Start where your curiosity points.

Distributed systems · authorization · agents

Can consent become a type system for agents?

Permissions are usually the last uninteresting thing in a system. A checkbox, a string in a database, maybe a banner that somebody clicks past. That is fine until you give a machine permission to make a plan.

Read the essay →
Cryptography · retrieval · personal computing

How do we retrieve private memory without handing over a life?

Memory is useful precisely because it is personal. That is also why treating it as an API payload is a bad habit.

Read the essay →
LLM systems · information retrieval · evaluation

What does it take to trace a grounded answer?

When a search answer is good, nobody asks how it arrived. When it is wrong, slow, or suspiciously certain, that is the only question that matters.

Read the essay →
MCP · capability security · programming languages

Can MCP tools carry security semantics?

A tool can look harmless in a sentence and still send an email, export a record, or change a customer’s state. Calling it “search” does not make it read-only.

Read the essay →
Systems · edge AI · scheduling

When should an agent leave the device?

The interesting question is not whether the cloud is fast. Of course it can be fast. The interesting question is whether the job deserves to leave the machine sitting in front of the person.

Read the essay →
Privacy engineering · metrics · distributed protocols

Can you measure usefulness without watching people?

The easiest way to learn what people do is to watch them. That is also exactly the habit a private-agent company has to break.

Read the essay →

Bring a better experiment.

If the argument is weak, the next useful thing is not applause. It is a tighter test, a clearer threat model, or a smaller working prototype.

Read the research notesJoin Discord

🤫 One is made by Hushh Technologies Corporation, an independent company. We name the hardware and clouds One runs on to say where it runs. None of them endorse us, and we call a company a partner only once the agreement is signed.