You documented it. You’re sure of it. But now — mid-client call, technician waiting — you can’t find it.
This is one of the most common complaints about IT documentation platforms: not that information isn’t captured, but that it can’t be retrieved fast enough to be useful. Weavestream’s search is built around that problem specifically. Here’s how it works.
Full-Text Search Across Everything
Weavestream indexes all of your documentation content — articles, asset records, and uploaded files — into a PostgreSQL full-text search index backed by native tsvector columns and GIN indexing.
That means results come from a real relevance model, not a simple LIKE '%keyword%' query. Titles are weighted more heavily than body content. Partial matches work. And because the index is maintained incrementally by a background worker as records change, results are always current without any manual reindex step.
What gets indexed:
- Articles — title and body text (rich text is converted to plaintext before indexing)
- Assets — asset name and all text field values from custom layouts
- Uploads — filename and attached notes
If the information is in Weavestream, search will find it.
Scoped to What You Can See
Search results are always filtered by the requesting user’s tenant access. A technician scoped to three clients cannot surface results from the other twenty. Client portal users cannot find records that have been flagged as internal-only.
This is enforced at the query level, not in the UI — so it applies consistently whether someone is searching from the main interface, the command palette, or the API.
The Command Palette
Press Cmd+K on macOS or Ctrl+K on Windows and Linux anywhere in the admin interface to open the command palette.
The command palette searches articles, assets, and companies in real time as you type. No need to navigate to a search page, select a category, or hit enter. It’s the fastest path from a question to an answer — which matters a lot when you’re on a call and need to pull up a network diagram or a set of credentials in a hurry.
Each user can tune the command palette to match how they work:
- Scope — search only the current tenant, or search across all tenants you have access to
- Depth — recent items only (fast, contextual) or comprehensive search (everything in the index)
For technicians who live in a single client environment all day, scoped search reduces noise. For engineers who move between clients constantly, cross-tenant search saves the extra step of switching context.
Why PostgreSQL, Not a Search Service
A lot of hosted documentation platforms rely on external search infrastructure — Elasticsearch, Algolia, or similar — which means your documentation content is indexed somewhere outside your control.
Weavestream keeps search inside your Postgres instance. There’s no external service to configure, no data leaving your environment, and no per-query pricing. The tsvector approach scales well for documentation workloads — thousands of assets and hundreds of articles — without any operational overhead.
For MSPs running self-hosted infrastructure, “fewer moving parts” is a genuine operational benefit. One less service to monitor, one less integration to break.
Putting It Together
The practical effect of solid search in an IT documentation platform is compounding. Technicians stop recreating documentation that already exists because they couldn’t find it. Onboarding new staff gets easier because information is actually discoverable. Client calls get shorter because the answer surfaces in three keystrokes instead of thirty seconds of folder navigation.
Weavestream’s search isn’t a feature you’ll spend much time thinking about — it’ll just work. The command palette in particular tends to become invisible in the best way: something people use dozens of times a day without noticing because it never gets in the way.
Try It
If you’re already running Weavestream, hit Cmd+K and start typing an asset name or a client’s network range. The command palette is on by default.
Not yet on Weavestream? The quickstart guide gets you running with Docker Compose in a few minutes. Self-hosted IT documentation search, under your own roof.