Homebase Building Blocks

What a work homebase needs.

A work homebase is built from a small set of building blocks. Each one addresses a failure mode in current AI products: amnesia, scattered context, unsafe access, weak verification, static interfaces, and users who do not yet know how to delegate.

How work becomes reusable

Intent becomes reusable work.

Memory returns to the next request, so every finished task becomes starter context.

01 Intent

Work Request Parser

Turns a messy ask into fields, assumptions, and a success shape.

Output: Parsed brief

02 Context

One Workspace

Pulls files, prior outputs, notes, and preferences into the task.

Output: Warm context

03 Boundary

Permission and Delegation Layer

Sets what the agent can read, edit, run, share, or ask about.

Output: Safe scope

04 Execution

Dynamic Tools

Creates the temporary surface, script, or workflow the work needs.

Output: Live tool

05 Inspection

Verification Loop

Shows sources, diffs, assumptions, and risks before approval.

Output: Reviewable result

06 Reuse

Accumulating Artifacts

Saves the output, procedure, and corrections as reusable material.

Output: Starter context

Memory return

The context gets richer.

Procedures, preferences, artifacts, and review patterns flow back into the next ask.

Prior outputsSaved procedureUser preferenceReview pattern

Every completed task should make the next task easier.

parse

messy askfieldswork order

Work Request ParserMainstream users ask for outcomes in messy language. Agents need a clearer task shape before they can gather context, ask for permission, or produce useful work.Work Request Parser is the front door of the work homebase: it converts a hunch into a task the system can actually execute.Open primitive

context

sourcesworkspacehistory

One WorkspaceChat sessions and project folders fragment work. Users keep rebuilding context because the product is not where the work lives.One Workspace is the equivalent of the software repo for knowledge work: the place where context, tools, artifacts, history, and future work can compound.Open primitive

reuse

briefrubricfuture task

Accumulating ArtifactsAI outputs are often disposable. Users copy them out, lose them, or recreate them later.Accumulating Artifacts turns AI output from disposable text into reusable workspace material that can be inspected, adapted, and called back into future tasks.Open primitive

surface

requesttooloutput

Dynamic ToolsHeavy GUIs freeze assumptions too early. Raw chat cannot always show the work clearly.Dynamic Tools let users get software shaped to the task without becoming software developers.Open primitive

inspect

sourcememorycontrol

Visible MemoryPersistent memory is powerful, but hidden memory creates distrust. Users need to see the system's assumptions before they can rely on it.Visible Memory makes personalization inspectable. The user can see what the system knows, where it learned it, and why it used that context.Open primitive

approve

readeditsend gate

Permission and Delegation LayerAgents need access, but access creates fear. Delegation only works when the user knows the boundary.The Permission and Delegation Layer turns security boundaries into product surfaces that help users feel in control.Open primitive

review

answersourcesrollback

Verification LoopKnowledge work usually lacks tests. The human remains the verifier, but most AI products make review harder than it should be.Verification Loop gives humans the review surface that knowledge work needs when there is no unit test suite.Open primitive

learn

democollabhandoff

Built-In TeachingUsers do not automatically know what to delegate or how to judge the result.Built-In Teaching makes the product improve the user's judgment, not just finish a task.Open primitive