Search
Documentation

← Documentation

Content types

Slidura supports seven content types that go beyond simple text fields. Each type is a reusable asset you author once and place into any number of builds via a short sentinel reference — for example @chart:<id> or @code:<id>.

Shared ideas

Media library and scoping. Content items live either in your personal media library (origin library) or scoped to a specific build or template. Build-scoped items are archived with their build; library items persist and can be reused across templates. The unified ref search (/refs/search) finds items across all origins and ranks the open build’s own items first.

Sentinels in fills. To place a content item into a build, set its sentinel as the value of the matching field in your fill.json. Each type has its own sentinel prefix so the filler knows which engine to invoke at build time.

Pure compute at build time. All rendering — chart data patching, table reshape, diagram transcode, code animation — happens inside the build job on Slidura’s servers. There is no per-asset render step after creation; the template’s design drives the output style wherever possible.

Shared data-grid editor. Charts and tables share the same spreadsheet-style grid widget in the web UI: editable headers, add/remove rows and columns, and a paste-to-import escape hatch for CSV or JSON bulk entry.

API and MCP surfaces. All major content types expose a REST endpoint under /api/v1 and an MCP tool. Agents can create assets, retrieve their sentinels, and drop those sentinels straight into a submit_build fill without touching the web UI.

Content types