> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vexa.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# A labelled, versioned evaluation set for meeting-to-task work

> A labelled, versioned evaluation set that scores how well agents turn meetings into tasks, so a change to that path is measured rather than argued.

**Is Vexa's meeting-to-task quality scored against a labelled set?**

**Today:** No. Evaluation exists at the module and contract level; there is no labelled set for meeting-to-task work.

**Sponsorable:** yes — The set, its labels, and the scoring.

***

*Score the meeting-to-task path instead of arguing about it.*

|                 |                                          |
| --------------- | ---------------------------------------- |
| **Status**      | **sponsorable**                          |
| **Size**        | M                                        |
| **North star**  | 4 · Agents that act on meetings, safely  |
| **Measured by** | verified delegated authority and tracing |

## What it does

* A labelled set of meetings and the tasks they should produce.
* Versioned, so a score is comparable across changes.
* Scored, so a regression is visible.

## What exists today

Evaluation exists at the module and contract level. There is no labelled set for meeting-to-task work.

## What this adds

The set, its labels, and the scoring.

## How sponsorship works

You fund the item. It moves to the front of the queue. You co-write the acceptance criteria, so "done" means done on your meetings. Your name goes on the release note. Sponsorship buys **the order of the queue** — never exclusivity, never a private build, and never a date the work has not earned. The fee basis is per item or per stage, agreed in writing before work begins.

## Two doors

<Columns cols={2}>
  <Card title="Sponsor this" icon="handshake" color="#0D9373" href="https://cal.com/dmitrygrankin/web?utm_source=docs&utm_medium=roadmap&utm_campaign=sponsor&utm_content=evaluation-set" horizontal arrow cta="Book fifteen minutes">
    Fifteen minutes. Bring the constraint you are under.
  </Card>

  <Card title="Watch this" icon="envelope" color="#0D9373" href="mailto:dmitry@vexa.ai?subject=Roadmap%20update%3A%20A%20labelled%2C%20versioned%20evaluation%20set%20for%20meeting-to-task%20work&body=watch" horizontal arrow cta="Email me">
    One email when it ships. The mail is already written — send it as it is.
  </Card>
</Columns>

## Questions

<AccordionGroup>
  <Accordion title="Is Vexa's meeting-to-task quality scored against a labelled set?">
    No. Evaluation exists at the module and contract level; there is no labelled set for meeting-to-task work.
  </Accordion>

  <Accordion title="What would sponsoring this add?">
    The set, its labels, and the scoring.
  </Accordion>

  <Accordion title="What happens when it ships?">
    It ships upstream under Apache-2.0, to everyone, with the sponsor named on the release note.
  </Accordion>
</AccordionGroup>

***

*Ships upstream Apache-2.0 when sponsored.*