WPIntell

Source evidence

Filtering posts during Import

MxChat – AI Chatbot & Content Generation for WordPress · support · 2026-04-02T01:49:00+00:00

mixedsentiment
mediumseverity
0.78relevance
1replies
Evidence linked to opportunitycommercial context

Proof Health

Open evidence

Commercial opportunities need traceable source links before they are treated as build-worthy.

7 / 29 rows with source links

24.1% of this page's analysis has direct source links.

0 build-decision rows missing links

0 rows here require auditable proof before promotion.

22 rows with no attached evidence

0 rows have source counts but still need direct links.

Conversation

support
ldeschenes unresolved
Hi Maxell, I’ve always found it valuable for developers to see how their software is used in real-world scenarios, so I wanted to share our use case. We manage a homeowners association website composed of three main content types: Pages (~250) Posts (~1,400 announcements over the past 10 years) Rules and Regulations (6 PDF documents) Our objective is to import these into the knowledge base for residents, with some filtering: Pages : Only those accessible to residents should be included (excluding Property Manager only content). Posts : Since announcements are often repetitive, we limit the import to the past year to reduce duplication. Rules and Regulations : These are handled separately via the PDF sitemap import. For Pages and Posts, I’m using the “WordPress Content” importer. To restrict the imported posts, I implemented the mxchat_before_process_post hook with a custom snippet. See below: add_filter(‘mxchat_before_process_post’, function($post, $bot_id) { // For pages only: skip if "gestion" is in the URL if ($post->post_type === 'page') { $permalink = get_permalink($post->ID); if ($permalink && stripos($permalink, 'gestion') !== false) { return false; } } // For posts only: skip if older than 1 year if ($post->post_type === 'post') { if (strtotime($post->post_date) < strtotime('-1 year')) { return false; } } return $post; }, 10, 2); This works partially: the content itself is excluded as intended, but entries are still created with null content , which results in empty indexed items. I can work with that, but would it be possible to adjust the logic so that entries with null content are not indexed at all ? Alternatively, a dedicated hook to explicitly exclude posts from indexing might be a cleaner approach. I look forward to your thoughts on the best way to handle this. Louis The page I need help with: [ log in to see the link] Hey @ldeschenes , Thank you for providing this – it’s very insightful. Reviewing it to see how this can be improved and in the next update. I’ll be in touch soon! Thanks, Maxwell

Comments

1 shown
m4xw3ll 2026-04-10T01:25:00+00:00

Hey @ldeschenes , Thank you for providing this – it’s very insightful. Reviewing it to see how this can be improved and in the next update. I’ll be in touch soon! Thanks, Maxwell