> ## Documentation Index
> Fetch the complete documentation index at: https://docs.smolboard.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Public feeds

> Schedule, speakers, and a subscribable calendar — no credentials.

Three anonymous endpoints back every public surface. They are the safest place
to start, and the only ones you can call from a customer's browser.

## Schedule

```bash theme={null}
curl -X POST https://www.smolboard.app/api/fn/getPublicSchedule \
  -H 'Content-Type: application/json' \
  -d '{"orgSlug":"ai-engineer","eventSlug":"ai-engineer-sandbox"}'
```

Returns `event` (including its IANA `timezone`), `rooms`, `tracks`, and
`sessions`. Two rules govern what appears:

* The organizer must have **published** the schedule. Before that, `sessions` is
  empty and `published` is `false`.
* Each session's content must be **approved**. Editing a title or abstract
  re-drafts it, so it leaves the feed until an organizer approves the change.

Session times are instants. Render them in `event.timezone`, not the viewer's —
a talk happens when it happens regardless of who is reading.

## Speakers

```bash theme={null}
curl -X POST https://www.smolboard.app/api/fn/getPublicSpeakers \
  -H 'Content-Type: application/json' \
  -d '{"orgSlug":"ai-engineer","eventSlug":"ai-engineer-sandbox"}'
```

Deliberately narrow: name, tagline, bio, company, job title, headshot. No email
addresses and no user ids.

## Calendar

```
GET /{orgSlug}/{eventSlug}/calendar.ics
GET /{orgSlug}/{eventSlug}/calendar.ics?sessions=<id>,<id>
```

A `text/calendar` feed of the published schedule, subscribable in any calendar
app. Pass `sessions` to narrow it to one attendee's picks. Unknown ids match
nothing rather than erroring.