AI Personal Assistant
Triages the inbox and drafts what you would have written
- 3output fields
- A structured result, not a paragraph you have to re-read
- ≤7steps per run
- Bounded work: at most 6 tool calls, enforced by the runtime
Read-only by design. This agent has no path to sending, spending, publishing or committing anything. Where that limit is the product, removing it would remove the reason to trust the output.
Overview
Reads what has arrived, sorts it by what actually needs you, and drafts the replies you would have written anyway.
Nothing is sent and nothing is scheduled without you: every outbound action is a write action held for approval by default.
Benefits
Daily
Inbox in minutes
Triage and drafts ready before you start.
Consistent voice
Drafts match how you actually write.
Flags what should not be answered
Phishing attempts and unusual payment requests are separated out rather than drafted a polite reply, which is the one message where a fast answer is the wrong one.
What it does
Priority triage
Sorts by what needs you, not by arrival order.
Drafted replies
Written in the voice you supply.
Day preparation
Summarises what the day requires.
Nothing sent silently
Every outbound action waits for approval.
What this changes
What this changes about your week
- It takes over
- The scheduling, chasing and small admin that fills the gaps in a day.
- Done today by
- The person whose time is most expensive.
- How often
- Continuously.
- Typical manual effort
- Admin takes an hour or two a day from whoever is least suited to it.
Why Productivity in particular. The work is trivial and the person doing it is not. That mismatch is the whole argument.
Runs unattended
Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.
The same standard every time
The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.
It cannot act on its own
AI Personal Assistant has no path to sending, spending or committing. That limit is why its output is safe to act on.
How it runs
This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.
- 1InputValidated against the agent input schema
- 2Modelclaude-sonnet-5 with automatic fallback
- 3Toolssummarize_text, extract_structured_data, send_email
- 4PermissionsRuns READ actions automatically; anything higher waits for approval
- 5OutputStructured result matching the output schema
- 6LoggingTokens, duration and every step recorded
Demo
Demo dataIllustrative sample output.
Input
{
"voice": "Direct, warm",
"messages": [
{
"body": "…",
"from": "supplier@example.com",
"subject": "Invoice query"
}
]
}Output
{
"drafts": [
{
"body": "…",
"subject": "Re: Invoice query"
}
],
"priorities": [
{
"why": "Blocks a payment run",
"subject": "Invoice query",
"priority": "high"
}
]
}Requirements
Integrations
Before you start
- Message text to triage
- Connect an email provider if you want an approved draft sent
- Available on every plan
Use cases
Morning triage
Start the day with a sorted inbox.
Meeting preparation
Summarise context before a call.
Pricing
Free
Free to deploy. Runs are metered against your plan allowance.
Runs consume your plan allowance for agent executions and tokens. See plan limits
FAQ
Can it send email for me?
Only after you approve. Sending is a write action and the default risk ceiling holds it for approval.
Does it read my mailbox?
No. You give it the message text for the batch you want triaged. There is no mailbox connection, so nothing is read that you did not hand it.
Who pays for the AI usage?
You do, on your own key, billed by your provider. Cost follows how many messages you put through it, which makes a daily triage batch easy to predict.
Reviews
Reviews open once customers have run this agent.
Need this agent to work differently?
Tell us what to change and our team will scope a customised version for your business.
Customize this agent