Inspect the public preview

What a Former WhichAI Intake Prototype Returned for One Frozen Task

An inspectable historical capture of a former WhichAI intake prototype, including the exact input, deterministic output, screenshot, method, and limitations.

By WhichAI. Published 2026-07-12. Updated 2026-07-14.

Methodology: Editorial synthesis of workflow design patterns and implementation constraints. Public control references provide context, not proof of a deployment or legal advice. Where a versioned evidence pack appears, its evidence class, method, and limitations govern what the artifact can support. Read the full method. Report a correction.

Built for

Prospective WhichAI users evaluating self-serve workflow planning

The decision

Determine what the archived prototype revealed, what it did not provide, and whether the current task belongs in paid self-serve research or a Solutions diagnostic.

Answer first

For the frozen weekly KPI task, the public preview returns a category, diagnosis, workflow decision, first tool role, human boundary, and first move. It does not return the authenticated researched report, configure an account, connect systems, or prove an operating result.

Self-serve workflow planner

Start with this article's task

For Prospective WhichAI users evaluating self-serve workflow planning. Start a brief for this task: Determine what the archived prototype revealed, what it did not provide, and whether the current task belongs in paid self-serve research or a Solutions diagnostic.

Start this brief

The capacity leak

What the team is doing before anyone calls it a systems problem

Headcount pressure rarely starts with one giant task. It starts when ordinary work is split across inboxes, tabs, handoffs, and undocumented judgment calls. These are the signals to map first.

SIGNAL 01

Product claims are easier to evaluate when the exact input and complete visible output are preserved together.

SIGNAL 02

A suggested tool role is not proof that a vendor account, plan, or integration fits the task.

SIGNAL 03

A deterministic preview can frame the decision without supplying the researched full report.

SIGNAL 04

Readers may infer account setup or implementation unless the boundary is explicit.

The implementation

The system should prepare the decision, not pretend the decision disappeared

A complete implementation connects the intake, context, transformation, review, and record. The output of one stage becomes the controlled input to the next. A human owns the exceptions and the final consequence.

Stage	Current drag	System responsibility	Human responsibility	Evidence kept
1. Freeze the task	The walkthrough uses an underspecified marketing example.	Publish the exact task, source path, product surface, and capture date.	The task owner checks that the example describes bounded recurring work.	Input, source path, product surface, date, and exclusions.
2. Capture the full preview	Only selected screenshots or claims are shown.	Preserve every field returned by the deterministic preview and a browser capture of the same task.	The reviewer checks that the JSON and visible interface agree.	Complete JSON, screenshot, timestamp, and checksums.
3. Check the narrow tool label	A suggested tool role is treated as a complete vendor decision.	Check only the named capability against current official documentation and list everything the check does not establish.	The buyer verifies account, plan, security, integration, and task fit before use.	Official URLs, access date, bounded claim, and exclusions.
4. Mark the operating boundary	The preview is presented as a finished implementation or researched report.	List the access, mappings, controls, failure tests, vendor checks, and local evidence still required.	The prospective owner decides whether to continue, investigate, or stop.	Boundary notes, unknowns, owner, and next action.
5. Offer routing	Every reader receives the same sales path.	Route known bounded tasks to self-serve and cross-system staffing or risk problems to Solutions.	The buyer chooses the path based on the visible decision rule.	Routing criteria, CTA, assumptions, and inquiry context.

What the human keeps

The goal is not zero humans. It is zero avoidable preparation around the judgment only a responsible owner should make.

A task owner checks that the frozen input reflects bounded recurring work.
An editor verifies that the published files match the current preview code and labels every boundary.
The buyer decides whether the remaining work fits paid self-serve research, Solutions, or no project.

Controls before volume

A workflow is not ready because the happy path worked once. It is ready when access, review, fallback, and evidence are explicit.

Publish the exact input, complete archived output, source path, and capture date.
Keep the narrow official documentation check separate from account, plan, integration, security, performance, and task-fit verification.
Do not present the archived prototype as the authenticated researched report or a current visitor entitlement.
Do not claim account creation, automatic connection, completed implementation, or customer outcomes.

Control references

National Institute of Standards and Technology: AI Risk Management Framework Federal Trade Commission: Operation AI Comply

The scorecard

Measure capacity, not activity

A system can produce more messages and still make the operation worse. Measure movement through the workflow, the quality of review, and the load that still reaches a person.

Capture completeness

Published fields and screenshot match the frozen preview output.

Boundary clarity

Missing research, verification, configuration, testing, and implementation work is explicit.

Capability-check scope

The narrow documented tool capability is separated from every unverified decision.

Next-action clarity

A buyer can identify paid self-serve research, Solutions, or no-project next step.

Inspectable original evidence

Archived intake prototype capture

This pack preserves the deterministic intake prototype that WhichAI previously showed for one frozen reporting task. It is historical evidence, not a current visitor entitlement or the paid researched workflow report.

Evidence v1

2026-07-14

Archived intake prototype

Method

Freeze the exact task, source path, and timestamp.
Run the public preview interface from the current source contract.
Check the published JSON against the function in an automated test.

Limitations

The archived prototype used local rule-based pattern selection and did not call a research provider.
No vendor account, integration, paid workflow generation, or customer outcome was tested.
The dated source check verifies only two narrow documented n8n capabilities. It does not verify an account, plan, configuration, integration, security posture, performance, pricing, or fitness for this task.

observed product output · Text

Exact frozen input

The complete task and source path used for the capture.

observed product output · JSON

Archived deterministic output

The complete former prototype output with no selected excerpts.

observed product output · PNG

Archived interface screenshot

A browser capture of the frozen task rendered by the former intake prototype.

official source verification · Markdown

Dated vendor capability check

Primary n8n documentation for the narrow capability label used in the preview.

observed product output · Markdown

Annotations and boundary

Field-by-field notes explaining what the output does and does not prove.

What a fake implementation looks like here

These patterns create an AI demo while leaving the labor, risk, and accountability in the same place.

Selecting only favorable prototype fields.
Treating a narrow documentation check as vendor or integration verification.
Presenting the archived prototype as the current paid researched report.
Implying that self-serve configured or connected the stack.

Two ways to act

Use the path that matches the decision

Task-specific workflow brief

Plan this recurring task.

Start with this task draft, then complete the three-question brief:

Prepare the weekly KPI dashboard from three approved source exports, explain material changes, link each figure to its source, route missing or conflicting values to a reviewer, and produce an approval-ready summary with a complete evidence record.

Choose a paid plan after reviewing your brief. WhichAI creates a plan and does not set up tools or accounts.

Start the brief

WhichAI Solutions

The workflow is becoming a company problem.

Use WhichAI Solutions when the walkthrough exposes custom system access, staffing pressure, sensitive data, unresolved risk ownership, or the need for a controlled pilot.

Bring one bottleneck. We map the work under it, separate consequential judgment from mechanical drag, and decide whether the next move is a hire, a tool, or a rebuild.

See company solutions

Questions

What operators ask before they build

What does this evidence show?

It shows the complete output of a former deterministic intake prototype for one frozen reporting task, the visible browser rendering, and a narrow documentation check for the suggested tool role.

Is this the current paid workflow report?

No. The authenticated report is a separate researched product surface and is not represented by this historical evidence pack.

When should the buyer choose Solutions?

When the task is really a staffing, backlog, cross-system, risk, or implementation problem rather than a request for a blueprint.

Primary references

Controls should come from the specific operating environment

These are broad public control references, not article-specific evidence, vendor endorsements, or legal advice. Validate the current rules, contracts, system configuration, and organization-specific risk before deployment.

National Institute of Standards and Technology

AI Risk Management Framework

A voluntary framework for mapping, measuring, managing, and governing AI risk.

Accessed 2026-07-14

Federal Trade Commission

Operation AI Comply

Enforcement examples showing why AI performance and substitution claims need evidence.