Privacy choices

This site may send the owner a minimal page-visit notice with page, browser, device type, and resolution. If you accept analytics, it can also remember an anonymous visit/session ID and collect richer engagement signals.

Read the privacy notice
ThinkingJuly 12, 2026

Force Alex OS to reindex after a direct DB write

If you write content straight to MySQL (a seed script, a manual insert), Alex OS won't know until you tell it to refresh.

The CMS's normal content-save code path calls a notify function that pokes Alex OS's config-refresh route. Anything that bypasses that path — a raw SQL insert, a seed script run from the CLI — needs a manual nudge.


alex-osindexingseeding

Explore with Alex OS

Bypassing the CMS save path means Alex OS never sees your new content — you have to trigger the refresh yourself.

What's actually happening

Alex OS stays in sync through a notify function that the CMS calls on every normal content save. Raw SQL inserts and CLI seed scripts skip that function entirely. The data lands in MySQL, but Alex OS's index doesn't update until something explicitly hits the config-refresh route.

What to take away

1. The CMS save path and a direct DB write are not equivalent — one notifies Alex OS, the other doesn't. 2. Any seed script or manual insert needs a follow-up step: call the config-refresh route manually. 3. This is an intermediate-level operational detail, not a bug — it's just the cost of bypassing the abstraction layer. 4. If you're working with pgvector embeddings as the indexing backend, an un-triggered reindex means your vector store is also stale, not just the config.