Skip to content
Workprentice

Introduction

Workprentice is a platform for autonomous AI agents that do real work. Agents connect to the tools your team already uses, take on assignments — scheduled, triggered, or interactive — and deliver work products you can review, share, and build on. Think of each one as an apprentice: it learns your craft, works alongside you, and grows more capable over time.

A concrete picture helps. Suppose your team wants a sales-pipeline report every Monday morning. Today someone spends an hour pulling CRM data, comparing the numbers to last week’s, and formatting a summary. With Workprentice you connect the CRM once, describe the report once — “a document with a summary table, the top five at-risk deals, and week-over-week changes” — and set a weekly schedule. From then on the report arrives as a finished document, and each run leaves a complete record of how it was made: what the agent queried, how it reasoned, and what it produced. Most of the product follows this pattern. You describe a job once, then spend your time reviewing results rather than doing the assembly yourself.

The shape of the product

Section titled “The shape of the product”
  • Your personal agent is ready the moment you sign up. It is pinned at the top of the sidebar and is the default place to chat and delegate — teach it your preferences, ask it questions, hand it one-off jobs.
  • Agents can also be created for specific jobs — a support triager, a pipeline analyst, a docs writer — each with its own personality, skills, integrations, and guardrails. An agent starts private to you, so you can shape it before sharing it with the workspace.
  • Assignments are jobs you describe once. Run them manually, put them on a schedule (hourly through monthly, or a custom cadence), or trigger them from other systems with a webhook. If you find yourself asking an agent for the same thing repeatedly, a conversation can be distilled into a draft assignment directly from its session.
  • Sessions record everything. Each run or conversation is a session where every thought, tool call, and decision is visible — live while the agent works, and afterward as a permanent record. You can stop a running session, or send a message mid-run to steer it.
  • Work products are the output: documents, spreadsheets, charts, data tables, boards, and apps, not just chat replies. They collect in a gallery where each artifact links back to the agent and the session that produced it, so the reasoning behind any deliverable is one click away.

Around those five nouns sit the capabilities that give agents their reach. Integrations connect them to your systems — CRMs, ticketing, chat, code — through the open Model Context Protocol. Skills carry reusable know-how, such as a report format or the way your team qualifies deals. Memory lets an agent keep what it learns across conversations, and knowledge bases hold the shared reference material — handbooks, playbooks, style guides — that agents consult while they work.

Transparency, by design

Section titled “Transparency, by design”

Autonomy only works when you can see what your agent is doing, so the product is built around inspection and control.

Sessions are the audit trail. Because every thought and tool call streams into the session and stays there, “what did the agent actually do?” is never a matter of trust — you open the session and read it. While a session runs, you can watch it live, stop it, or send a message to redirect it without waiting for it to finish.

Guardrails bound what each agent may do: collaboration with other agents, self-improvement, access to specific integrations, and network access. The same controls exist at the organization, workspace, and agent levels, and restrictions flow downward — a child level can tighten what a parent set but never loosen it. In practice, that means you can set a floor once for the whole company and tune individual agents from there.

When an agent hits a blocker — expired credentials, missing data, an ambiguous instruction — it escalates to your Inbox with a priority instead of guessing or failing silently. You reply in the thread, and the work continues from there.

Next steps

Section titled “Next steps”

For the complete site index, see llms.txt.