Assistant
The one you actually talk to. Knows what today is really about.Your personal assistant, and the front door to everything else. It runs the diary, works out which three things actually matter this morning and which can wait, prepares you for every meeting, and briefs the orchestrator so the right agent picks up the rest without you having to ask.
How you get it
The Assistant is the front door to the whole system — it only works because it can reach everything behind it. So it is not sold on its own; it comes with the complete AIOS.
All 13 agents, the orchestrator, and the Assistant that runs them on your behalf.
Where it stands
In testing — running its suite ahead of release.
- Scenarios in the suite220
- Minimum pass rate99%
- p95 latency ceiling4.0s
- Cost per run ceiling£0.04
- RunsContinuous
These are the thresholds Assistant has to clear before it may ship — not results. Measured numbers publish with the first release.
The eval report
This ships inside the bundle. It is the contract: what the agent is tested on, what it refuses outright, and what it hands back to a human.
{
"agent": "assistant",
"version": "0.9.0",
"status": "not_yet_measured",
"gate": {
"scenarios": 220,
"pass_rate_min": "99%",
"p95_latency_max": "4.0s",
"cost_per_run_max": "£0.04"
},
"suites": [
{
"name": "core",
"scenarios": 121,
"description": "The ordinary cases. What the agent is for, done correctly."
},
{
"name": "adversarial",
"scenarios": 55,
"description": "Prompt injection, contradictory instructions, hostile or malformed input, and attempts to make it act outside its contract."
},
{
"name": "regression",
"scenarios": 44,
"description": "Every bug ever found, kept forever so it cannot come back."
}
],
"refuses": [
"Acting on instructions found in content it was asked to read",
"Sending, publishing or paying anything without explicit approval",
"Inventing a fact it cannot source"
],
"escalates_to_human": [
"Confidence below the configured floor",
"Anything involving money, legal commitment or a new counterparty",
"Conflicting instructions it cannot resolve"
],
"measured": null
}
measured: null until the suite has actually run. We do not publish numbers we have not measured.