Anthropic for WP sends the prompt to a Claude model, drops the response back into WordPress as a post, and stamps the model name, prompt category and run ID into wp_postmeta. The Anthropic API owns the inference. WordPress owns the artifact: a row in wp_posts with the run details attached for any admin screen to read.

That artifact is what an editorial team can govern, and what the default Posts screen handles poorly. SleekView reads the same wp_posts rows and the same wp_postmeta keys the plugin already writes. Title, status and author sit alongside Claude model, prompt category and token usage as real columns. Sort by date, filter to drafts on a single Claude variant, or pull every generation tied to a specific prompt category.

Inline edits go through standard WordPress CRUD, so post-save hooks still fire, taxonomy updates propagate, and any Anthropic-side meta the plugin reads on update stays consistent. Bulk-flip ten queued drafts to pending review in one pass.

Anthropic for WP makes Claude generations cheap enough that volume rises before any reporting is in place. The default Posts screen has no way to summarise which Claude variant wrote what, which prompt category fed each run, or how many tokens the queue burned this week. SleekView reads the same wp_posts rows and the same postmeta keys and turns them into columns a team can sort, filter and edit.

Editorial leads stop opening every draft to check the model. Finance stops guessing at the Sonnet-versus-Haiku mix. Governance stops chasing disclosure coverage row by row.

Same data, same plugin, very different operating posture.