Know what your AI agents are doing. Prove it later.
When you run more than one AI agent, they start talking to each other — and today, nobody’s watching. Positif sits in the middle like a switchboard operator: every request between agents is permission-checked, logged, and loop-protected before it goes through. It never reads the messages — a phone bill, not a wiretap.
Your agents talk behind your back.
One AI agent asks another to do something. No one approved it, no record exists, and if something goes wrong you can’t reconstruct what happened. The more agents you add, the worse it gets — every agent wired to every other agent, one hijacked agent able to reach them all. Positif replaces that tangle with a single governed relay. Drag to see it.
Every request goes through the switchboard.
Positif checks each call is allowed, makes sure agents aren’t stuck in a loop calling each other, and writes it in a log that can’t be edited afterward. It reads only the outside of the envelope — who’s sending, who’s receiving — never the letter inside.
Place a call through the switchboard.
Pick a caller and a target. Watch the envelope move through the pipeline — and watch the ledger above catch the outcome. Notice what happens when a web-facing agent tries to reach something sensitive.
Runs on your own private network. Or let us run it.
Positif is yours to run — on your own private network (via Tailscale), where your keys and agents never leave. That is the whole point, and it is free forever. When you would rather not operate the control plane yourself, a managed option is coming. Either way you are never locked in: take it in-house any time, no export dance.
One half of a two-part fleet OS.
Positif is a sibling, not a standalone. RADLAB ships two organs of one body: Bourdon, the cross-agent memory, and Positif, the cross-agent routing. Together they are a fleet OS: what the fleet knows, and what the fleet is allowed to do. They share a philosophy — measured, payload-respecting, built in public — but each owns its branding. Positif keeps its own mark and its own sage; nothing is shared across the line.
Routes the fleet. Reads nothing. Proves everything.
Pre-alpha. Here’s what’s missing.
Naming the gaps is part of the pitch. Positif is built in public and changes weekly; treat anything below the line as a promise we have not kept yet.