Jev puzzle explanation coverage

Check a puzzle explanation against the supplied solution

Provide the fictional puzzle's solution facts and a proposed explanation. Distinguish complete reasoning from missing causal links or contradiction without relying on remembered puzzles.

puzzle explanation coverageLive API
Try an example

The explanation covers all three required links.

Source: API-collected workflow inspiration; original demonstration below

Source and example scope

This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Compare the proposed puzzle explanation with the supplied authoritative solution and required facts.

Decision rules and labels
CompleteIncompleteContradictoryReview
A real response from the Jev API

Edit any input to explore a different decision. No request is sent until you run it.

Your next decision

Not run yet

Choose an example or write your own input, then run it to see the ranked choices.

Results are not stored by this site. Your submitted text is processed by the API provider.

The decision

What the decision checks

The answer key is part of the input in this teaching example. The model compares a proposed explanation with that supplied mechanism instead of trying to solve an unseen mystery from memory.

An incomplete answer can be reasonable while still missing the link that explains the puzzle. Keep that separate from a statement that directly contradicts the solution.

Soupbase uses Jev within a lateral-thinking puzzle workflow. This original drawer-and-lamp puzzle demonstrates explanation coverage only; it does not copy a published puzzle collection or implement the full game host.

This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Illustration of selecting a caption token for emphasis in an edit timeline
Illustration of the decision flow.
How it works

Try the example in four steps

Use the sample to understand the rule, then test your own input for puzzle explanation review.

  1. Illustration of adding the text or evidence for a decision

    Add the information to check

    Provide the fictional puzzle's solution facts and a proposed explanation.

  2. Illustration of defining a question and its allowed choices

    Confirm the question and labels

    Open the decision rules and keep the intended distinction between Complete, Incomplete, Contradictory, Review. Edit the choices only when testing your own rule.

  3. Illustration of sending one input and comparing the returned choices

    Run the decision

    Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

  4. Illustration of reviewing a result and an uncertain item before using it

    Review the next step

    Review Review before using a result. The label is a decision about the supplied information, not an executed action.

Our API test

Recorded API observations

Recorded API observations
Original inputObserved labelLeading probability
All links explainedComplete94.0%
Power source onlyIncomplete100%
Wrong triggerContradictory99.0%
Paraphrased complete solutionComplete100%
No answer keyReview99.0%
Conflicting answer keyIncomplete58.0%

These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 5 of 6 results matched the authored fixture expectations. For Conflicting answer key, the expectation was Review; the observed choice was Incomplete.Verified

Understand the results

Puzzle explanation review: six recorded examples

The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.

Using the result in your application

The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.

Replay options

Inspect saved outputs or request a new decision

View the code example
Two independent commands
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json

# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.
Scope of this demo

Where this decision fits

Useful scope

  • Giving feedback on an original puzzle answer.
  • Testing whether a paraphrase covers the same mechanism.
  • Separating omitted explanation from contradiction.

Outside this demonstration

  • Declaring a solution correct without a consistent supplied answer key.
  • Presenting fictional puzzle reasoning as physical diagnosis of real equipment.
Example FAQ

Using this example

The input contract and these limits define what the result means.

Why provide the hidden solution to the model?

Coverage checking needs something authoritative to compare against. The player may not see the key in a real game, but the judging component needs it.

Must the answer use the exact required words?

No. A clear paraphrase can express the same mechanism. The example checks meaning rather than a keyword checklist.

Does the leading probability measure accuracy?

It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.

Will a result change an external system?

No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.

Test a different input

Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.