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

# Workload insights

> Coming soon: name the recurring workflows your workspace runs, then attribute requests to them and read a per-workflow summary. What is built and what is next.

Workload insights is for a workspace that runs the same operation over and over
and wants to see it as one thing: an extraction job, a summariser, an agent
loop. You name the workflow once, your requests say which workflow they belong
to, and the usage screen rolls them up by workflow instead of by key or model.

## What is built: the workflow registry

A workflow is a named, versioned record that belongs to the workspace. It
carries:

* a **name**, up to 80 characters, and an optional **description** up to four
  kilobytes, which is stored encrypted;
* a **declared type**, one of `extraction`, `summarization`,
  `classification`, `coding`, `question_answering`, `content_generation`,
  `agent_workflow` or `unknown`, or none;
* an optional **metric definition**: how you will judge a run of this workflow
  succeeded, as an application assertion or a human acceptance, with a version
  so the definition can change without losing history;
* an **archived** flag. Archiving clears the description and does not restore
  it on unarchive.

Every write bumps the workflow's version by one, and a revision names the
version it expects, so two people editing the same workflow cannot silently
overwrite each other. A workspace may hold 100 active workflows; archived ones
do not count.

Saving a workflow classifies nothing and records no consent. It is a label,
and the label does nothing until the next slices land.

The registry has a section on the Settings screen, where the owner names and
revises workflows and everyone else sees the list, and a set of endpoints on
the [management API](https://docs.hopscotchlabs.ai/manage/management-api.md). Both are owner-only to write and
open to every role to read.

## What is still to come

In the order the work is planned:

1. **Attribution.** A request header naming the workflow it belongs to, read
   on the way through and never forwarded to a provider. A label never changes
   which model serves the request.
2. **Summaries.** Request counts and charges rolled up per workflow, joined by
   request identity to the same records Activity shows.
3. **The usage screen.** The roll-up beside the existing by-key and by-model
   views.
4. **Tasks.** Bounded units of work inside a workflow, with their own scoped
   credentials.
5. **Deletion and retention.** Removing a workflow's observations on your own
   schedule.

Each of those is a story of its own, and this page gains a section as each one
ships. Nothing here promises a date.

## Next

> **[Monitor usage](https://docs.hopscotchlabs.ai/manage/monitor-usage.md)**
>
> The usage screen as it stands today, by key and by model.

> **[Management API](https://docs.hopscotchlabs.ai/manage/management-api.md)**
>
> Where the workflow endpoints live.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.