Skip to content
← The showcaseComplete free read
Public repository and hosted demo

Search with visible sources instead of generating an answer

A search interface lets Jev choose queries, sources, and time ranges, then ranks links without composing a synthetic answer.

Editorial workflow map
  1. Interpret the request
  2. Search selected sources
  3. Review ranked links
View the author’s original post (opens in a new tab)
Our interpretation · not the author’s screenshot or a reproduced result

What the repository shows

Jev selects a query, source set, and time window; Search1API retrieves results; Jev scores relevance. Filters remain editable and results retain outbound links.

Supporting documentation (opens in a new tab)

Why the output shape matters

The interface returns evidence candidates rather than a generated conclusion. A reader can change filters, inspect snippets, and open the underlying page.

What it does not establish

The repository warns that relevance percentages are model judgments and do not verify a result. Provider calls, caching, snippets, and dates introduce additional limits.

Use the same checks on another demo.

Read the complete free evaluation guide →