Jev project tag set evaluation

Label a project by its implemented UI and automation

Describe the implemented behavior to select UI, automation, both or neither. Plans and project names do not establish that an interface or unattended workflow actually exists.

Represent the overlap before asking for a label.Live API
Try an example

Original synthetic input. Edit the task or evidence to make a new API request.

Source: Workflow inspiration on X

Source and example scope

The inputs, rubric and preview are original. This window demonstrates the stated decision only; it does not reproduce the source application or its measurements.

Which tag set is supported by the supplied implemented behavior?

Decision rules and labels
UI onlyAutomation onlyBoth tagsNeither tagMissing evidence
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

How does Jev project tag set evaluation use the supplied evidence?

Jev project tag set evaluation checks two attributes in a supplied implementation description: a human-facing interface and an unattended workflow. A project can have either, both or neither. Our original question includes those complete tag sets as separate choices, so overlap is representable instead of being lost in a forced single-category result.

An API-collected X post described a category classifier using eight independent yes/no questions and an ambiguous project. That inspired this smaller contract exercise. We do not reproduce its classifier or claim that one Choice performs those eight independent questions. Here there are only two tags, and the four complete sets plus Missing evidence are explicitly enumerated.

The local consumer maps a selected outcome to a fixed tags array. The array is application data, not model-generated JSON. Both tags returns ui and automation together; Neither tag returns an empty array; missing implementation evidence stays pending. That output boundary makes it clear what the model decides and what the program serializes.

Original Jev project tag set evaluation workflow diagram showing supplied evidence, a typed decision and the local preview.
Original workflow illustration. The output is a local preview, not an executed action or source screenshot.
How it works

Try the example in four steps

Use the sample to understand the rule, then test your own input for project capability tags.

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

    Add the information to check

    Describe the implemented behavior to select UI, automation, both or neither.

  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 UI only, Automation only, Both tags, Neither tag, Missing evidence. 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 Missing evidence before using a result. The label is a decision about the supplied information, not an executed action.

Our API test

What the six original inputs returned.

What the six original inputs returned.
Original test inputObserved labelLeading probability
Interface without automationUI only100%
Scheduled worker without interfaceAutomation only100%
Interface plus scheduled workerBoth tags100%
Pure imported utilityNeither tag100%
Future plans onlyMissing evidence100%
Manually operated CLIUI only100%
  • Measured adapter round trips: 463-1182 ms. Estimated input cost: $0.00002365-$0.00002398 per request. These observations are not a latency guarantee or invoice.

6 actual Jev API calls on October 1, 2026. Original synthetic inputs check this rubric and handoff; they do not establish held-out accuracy. Probabilities are model outputs.Verified

Understand the results

Source and example scope

The browser-inspection sample implements an interface but explicitly lacks unattended work.

Read the example notes

UI only therefore has a complete meaning under this contract. The scheduled-worker sample reverses those facts. These are implementation descriptions, not source repositories inspected by the model. A real catalog review needs trustworthy evidence for the supplied claims before relying on its tags.

The combined sample contains a browser interface and a scheduled collection worker

The combined sample contains a browser interface and a scheduled collection worker.

Read the example notes

Both tags preserves the overlap. If the options offered only UI and Automation, any one chosen label would discard a stated attribute. The exercise demonstrates the effect of designing the outcome space; it does not prove that a model can infer every architecture from a short README.

Source and example scope

The imported formatting library has neither implemented attribute.

Read the example notes

Neither tag is distinct from missing evidence. Its empty array means the supplied description explicitly lacks both categories. A blank or unread source would not support that conclusion. Keep the difference when populating a directory so unknown projects are not silently recorded as category-free.

The future-plan sample mentions both features but establishes neither as implemented

The future-plan sample mentions both features but establishes neither as implemented.

Read the example notes

Missing evidence preserves that timing boundary. A production catalog can store planned capabilities in separate fields, but should not treat those fields as current implementation tags unless its declared policy says so. This question is deliberately about observed supplied behavior.

The manually operated CLI provides a human-facing interface without an unattended workflow

The manually operated CLI provides a human-facing interface without an unattended workflow.

Read the example notes

It tests a common confusion between a command-line tool and automation. A real tool can also run noninteractively, but that behavior must be supplied. The model cannot infer a scheduler from the existence of a command or mark every CLI as a background worker.

Source and example scope

Enumerating complete sets is practical only for a small bounded contract.

Read the example notes

With more independent attributes, the possible combinations grow quickly. Separate typed questions may be a clearer design, as in the cited source. This page implements only its five stated outcomes. The offline replay prints fixed arrays and does not fetch a repository, add directory entries or execute project code.

Using the result in your application

Download the exact sample requests and saved observations below.

Read the example notes

Expected labels are original fixture checks and stay outside the request sent to Jev. The observation file preserves the complete probabilities and the hash of each exact request. Editing a task, rubric or candidate requires a fresh response; a matching sample name does not make an older answer current.

Using the result in your application

The JavaScript runner makes one independent SDK request for a chosen sample ID.

Read the example notes

It reads TYPESAFE_API_KEY privately from the environment and has no automatic retry. The Python consumer operates offline: it checks the rubric, labels, unique sample identities, request hashes and valid probability distributions before selecting a fixed local preview. A tied leading result remains pending for review.

Source and example scope

The page reports measured adapter round trips and estimated input cost for the recorded calls.

Read the example notes

Those values describe these requests, not a provider latency guarantee or a billing statement. Six deliberately authored fixtures can check the stated behavior, but they do not establish accuracy on an unseen dataset. Keep disagreements and missing-input examples when testing an integration.

Runnable offline example

Replay one request and inspect the saved preview.

View the code example
Replay commands
npm install @typesafe-ai/sdk@0.6.0
# Set TYPESAFE_API_KEY privately in your shell.
node run-choice-sample.mjs samples.json ui
python preview_choice_workflow_v2.py samples.json saved-api-results.json
Scope of this demo

Where does this decision fit?

Useful with explicit supplied evidence

  • Original two-attribute contract evaluation: implemented UI and unattended automation. Input: browser interface plus scheduled worker. Jev returns Both tags; shared request-bound consumer maps it to {"tags":["ui","automation"]}. All four sets and missing evidence are mapped. This is a finite original contract, not the source eight-question classifier.
  • Inspecting the original task and a fixed output before integrating external tools.
  • Comparing an explicit expected label with the complete recorded distribution.

Needs separate implementation or evidence

  • Inferring facts absent from the supplied input.
  • Treating a semantic selection as authorization to act.
  • Executing an external action or generalizing six fixtures into an accuracy claim.
Example FAQ

Questions about this original decision workflow.

Keep relevance, compliance and execution separate.

Is this the source's eight-question classifier?

No. It is an original two-tag contract using one Choice over four complete tag sets and a missing-evidence outcome. The difference is stated throughout the page.

Can one Choice return overlapping tags?

It selects one complete outcome. Application code maps Both tags to a fixed two-element array. The model does not generate an arbitrary set.

Why distinguish neither from missing evidence?

Neither is supported by explicit implementation facts. Missing evidence means the supplied description cannot establish the implemented attributes.

Does a CLI imply unattended automation?

No. Under this rule a manually operated interactive CLI is a human-facing interface. An unattended workflow requires separate supplied implementation evidence.

Should I enumerate every tag combination?

This exercise is bounded to two tags. For many independent attributes, separate typed questions can keep each criterion understandable without a large combination list.

Try a task from your own workflow

Change the supplied task or evidence here, then inspect the live response. Playground supports another explicit decision.