No. SleekView Feedback reads directly from whatever table or post type the OpenAI plugin uses. You point it at the source, pick the columns for votes, status, category, author, and title, and the board renders. No ETL job, no sync, no duplicated data. Anything OpenAI writes shows up on the next page load.

Yes. SleekView ships with anonymous voting backed by cookies so public visitors can upvote prompts and drafts without an account. You can also require login if you want the board restricted to editors or paying members, and the same view handles both modes with a single toggle.

Each visitor gets a cookie scoped vote token per item. Logged in users are tracked by user ID. A built in rate limit caps how often a single IP can hit the vote endpoint, which keeps public boards honest without forcing a signup wall in front of casual readers.

Yes. SleekView accepts a WHERE clause when you wire up the data source, so you can scope the board to GPT 4o runs, GPT 3.5 runs, runs from a particular campaign, or any combination of meta fields. Different boards on different pages can use different filters.

Hallucination is just a category value on the row. You can write it into a meta key the plugin already understands or a dedicated column. Either way it shows up in the WordPress admin alongside the original prompt, so the prompt engineer who wrote the template can see the flag without leaving WordPress.

They write back to the source column, which means the plugin and any of your own queries can sort future jobs, retries, and bulk runs by that score. Several teams use the score to gate which prompts get more token budget, which makes the board operational and not just a vanity dashboard.

Both. SleekView ships as a Gutenberg block, an Elementor widget, a Bricks element, and a classic shortcode. Theme developers can also call the render function from PHP and pass a configuration array, so you can mount the board on any template without touching the page editor.

The view paginates server side and only loads the rows it needs to render the current page. Indexed columns stay fast even on long tables. For really big projects, scoping the board by campaign or date keeps both the query and the audience focused so the page feels snappy even at scale.