Sanity can model a playful, human-approved agriculture workflow, but the project behind this *Idiocracy*-inspired demo does not show that a picture-button interface is easier for non-programmers—or that workflow checks alone secure content. Its author, xbill, reported an agent-only pilot in which the stock Sanity Studio had the fastest median times on all three tested tasks. The human usability test was still future work.
What is the project?
In the *Idiocracy* premise, crops are failing because fields are watered with Brawndo. Joe’s plan is to water them with water. Xbill turned that joke into a Sanity-powered Department of Agriculture simulation: an agent proposes watering fields, a human Cabinet approves or rejects the proposal, and approved fields move through watering, growing, and harvest.
As an Amazon Associate I earn from qualifying purchases.
The project explores a practical content-management question beneath the satire: can a person interact with structured data through a purpose-built interface rather than editing documents or Markdown directly? It includes a picture-button “Kiosk,” the standard Sanity Studio, and a GitHub Markdown editing comparison.
How does the approval workflow work?
The author models the proposal as a process that changes stage, rather than relying only on a status field. The flow begins with a petition, advances to the Cabinet for a decision, and then branches: approval leads to watering, growing, and harvest; rejection leads to a dust-bowl outcome. Sanity’s Workflows overview describes the product’s process model in terms of definitions, instances, stages, activities, and effects.
#1 Best Overall
The agent can report on the situation, create a proposal, and submit it. It does not have an approval tool, leaving the decision to the human Cabinet. In the project implementation, vote tallies are computed with GROQ count(); a vote ID built from a citizen and field replaces that citizen’s previous vote for the same field. Those are details of this simulation, not universal Sanity voting behavior.
What did the implementation use?
Xbill’s September 18, 2026 article describes a specific setup, not a set of current prerequisites for building with Sanity. The reported implementation used a Sanity Growth-trial project, Sanity Studio 6, two workspaces on one schema, joes-plan v1, and Workflows packages at 0.33.0 early access. The front end used Next.js 16 and next-sanity; Sanity Functions and Cloud Scheduler handled runtime work.
The custom Kiosk uses picture-button inputs, while the stock Studio provides the standard editing surface and GitHub offers a Markdown-based alternative. Sanity’s App SDK overview describes a route for building custom content applications and workflows, but the project’s dated stack should not be read as a recommendation or a current compatibility guide.
Was the Kiosk faster than the regular Studio?
No—not in the reported pilot. The timings cover one participant, an agent labeled A1, completing three tasks three times on each interface. Xbill reports all 27 edits correct: nine per interface. Stock Studio had the lowest median time for each task.
| Interface | Water a field | Add a citizen | Amend a proposal | Correct edits |
|---|---|---|---|---|
| Stock Sanity Studio | 24.8 seconds | 16.8 seconds | 37.6 seconds | 9/9 |
| Picture-button Kiosk | 26.4 seconds | 25.6 seconds | 41.4 seconds | 9/9 |
| Markdown on GitHub | 37.2 seconds | 41.2 seconds | 46.2 seconds | 9/9 |
These are author-reported medians from the agent’s three runs per task and interface, not a general usability benchmark. The agent’s times include model decision time, and the interfaces were tested in the order Kiosk, Studio, then GitHub, so first-run learning may have affected results. The figures do not establish which interface is easiest or fastest for a human, particularly someone unfamiliar with programming.
Does Sanity Workflows enforce approvals securely?
No. Workflows can organize the approval process, but workflow stages, engine verdicts, and guards are not a hard security boundary against a client able to write directly to the Content Lake. Sanity’s access-control documentation says the Content Lake is the enforcement point. In practice, the project’s human approval step is meaningful as application workflow, but authorization must be enforced through dataset access controls—not assumed from the workflow path alone.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Has a non-programmer tested it?
Not in the article’s reported results. The only participant was the agent; xbill described testing the protocol with a person who had never used Git as the next measurement. The project therefore raises the non-programmer usability question but does not answer it.
Recommended Free Tools
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
