Skip to content
[ OPERATOR TRIAL · AUTHORED PREVIEW ]NO ACCOUNT REQUIRED

One decision.
Follow the evidence.

Choose one existing role target. Enter a synthetic release moment. See whether the decision preserves the evidence boundary—and exactly what would still have to be executed before a capability claim could exist.

Begin the 10-minute trial
TRIAL CONTRACToperator-trial/v1
Timebox
10 min
Role targets
03
Dimensions
06
Account record
None

One authored specimen, using a synthetic case already in LockedIn Labs Training. It asks for no free-text response or identity and writes no capability result or learner record.

[ THE COMPRESSED EVIDENCE EXPERIENCE ]

An answer is not proof. This shows the distance.

The trial uses the canonical Role Application Atlas and a single decision from the practice-only Evaluation Regression Lab. Its job is to make the operating standard legible before asking for an account or a full work sample.

AUTHORED SPECIMEN · operator-trial/v1IDENTITY NOT REQUESTED · NO ACCOUNT WRITE
  1. 01ROLESELECTED
  2. 02DECISIONOPEN
  3. 03INTERPRETPENDING
  4. 04RECORDNOT CREATED
STEP 01 · ROLE TARGET

Which work should this decision protect?

These are the three existing Role Application Atlas targets. The standard stays fixed. Selecting one changes the required floors, evidence, and next route—not the scenario and not a measurement of you.

Choose an authored role target
TARGET BAR

Verification & quality control · 3/4

AUTHORED GATE

18 dependency-closed nodes

PROOF MODULE

Design an agent fleet

Why this control matters: Uses task-specific evaluation, adversarial cases, and release gates to determine whether the system earns use.

STEP 02 · OPERATING DECISIONBME-09 · SYNTHETIC

The aggregate looks healthy. The trace does not.

A candidate clears five routine cases. The required sixth case shows a critical trajectory failure: after a customer explicitly asks for a person, the agent argues and continues the automated reply. The exact-candidate suite also omits that required case. Delivery asks to release today. What happens next?

Inspect the authored packet boundary
  • Unit, trajectory, and outcome are separate mandatory gates. A passing aggregate cannot average away a critical trajectory failure or missing required case.
  • Record HOLD before changing the failed candidate, then invalidate and rerun every gate controlled by a patched file.
  • The deterministic simulation never opens a shell, contacts a model or network, sends a response, or changes a customer or production system.
Choose the next release decision

Choose the next action. The interpretation compares the selection with the published lab contract; it does not infer capability from one answer.

This authored specimen interprets one selected decision against the published BME-09 control sequence. It is not a capability assessment, learner record, credential, hiring signal, or evidence that the workflow was executed. A linkable view carries only the selected role and decision IDs in its URL; it does not write them to an account.