Friday the 13th kindness challenge: a 13-clip recap in one render

Awareness Days suggests primary schools run a "Friday the 13th kindness challenge" for World Kindness Day 2026, where classes complete 13 kind acts and display the results. A school, club or office can turn that into a recap video: 13 clips of four seconds each make a 52-second video, one Timeline render ($0.10) and one caption job with 13 numbered cues ($0.20), $0.30 in total.

The idea and the date, Friday 13 November 2026, come from World Kindness Day 2026 on Awareness Days, read on 2026-10-12. The limits come from the Sume Timeline and video captions docs, read the same day.

Why is 13 slots a special case?

Timeline's default render.strategy is auto, which chunks a render past 12 segments. You can ask for single, but the docs say Sume refuses it above 12 slots with render_strategy_unsafe. So with 13 clips, leave the strategy alone. A second limit matters too: more than eight adjacent fades is refused as too_many_chained_transitions, so put a hard cut between some clips.

How do you build the body?

The script below builds the slots and the numbered cues and prints a summary; it runs with plain Python. Post the render object to POST /v1/timeline-1.0/plan first (it is unbilled), then to /render with an Idempotency-Key, and send cues to POST /v1/video-captions once the render is a finished artifact.

import json

ACTS = 13
SECONDS = 4
video, cues = [], []
for i in range(ACTS):
    start = i * SECONDS
    slot = {
        "source_url": f"https://media.sume.com/artifacts/artf_demo/act-{i + 1:02d}.mp4",
        "start": start,
        "duration": SECONDS,
    }
    if i and i % 2 == 0:
        slot["transition"] = {"type": "fade", "duration": 0.25}
    video.append(slot)
    cues.append({
        "text": f"Kind act {i + 1} of {ACTS}",
        "start": start + 0.3,
        "end": start + 3.5,
    })

total = ACTS * SECONDS
render = {"audio": {"mode": "silence", "duration_seconds": total}, "video": video}
print(len(render["video"]), "slots,", total, "seconds")
print(json.dumps(cues[0]))

What does it cost?

The 52-second render rounds up to one output minute, and the 52-second video is under the 60-second price band for a caption job.

Photograph acts that children and staff agreed to share, and leave faces out where you lack consent. The clips must be imported first, since Timeline refuses off-host URLs.