Turn a presentation plan into a teleprompter session
Turn a written video presentation plan into the recording session the camera screen runs: the teleprompter script sentence by sentence, the countdown, when to stop on silence, on the final word or on a stop phrase, and whether it may start by itself
Give plan (one compiled video presentation plan: id, title, medium video, execution.adapter capture.video, beats each with delivery speak and text, no open acquisitions, and the execution settings: countdown_seconds, auto_stop_seconds, silence_stop_seconds, silence_threshold_rms, minimum_speech_ms, stop_on_final_word, completion_grace_ms, completion_slack_words, script_end_quiet_seconds, stop_phrases, mirror_script, speech_language, follow_speech, allow_manual_nudge, completion_label), and optionally readiness (the setup pass's result for this plan), authorization (a grant: plan_id, authorized, authorized_at_ms, grant_until_ms) and nowMs (epoch milliseconds; the run's clock when left out). Answers the record-video.recording-session: session_id, title, subtitle, the stop and speech-follow settings with their defaults filled in, sentences (the teleprompter script), source_plan, requirements, plan_capabilities, readiness_state, authorization_state and auto_start, which is true only with a complete matching readiness and an unexpired grant for this plan. Refuses a plan that is not a video plan for capture.video, has open acquisitions (a gathering format), has no beats or a beat that is not spoken text, has a negative or non-numeric setting, and a grant that claims capture without fresh readiness or that does not match the plan and time. It reads only its input: it never opens a camera or microphone, never treats a plan as consent, records nothing and asks no AI to write a script. The screen's other work (the camera picture, countdown, live waveform, silence stop and speech-following teleprompter while recording, voice-only choice, and having the person's AI write the script for a format) needs a browser and is not done by this run.
What it asks for
Nothing: your AI starts it with your words.
Use it from your AI
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to turn a written video presentation plan into the recording session the camera screen runs: the teleprompter script sentence by sentence, the countdown, when to stop on silence, on the final word or on a stop phrase, and whether it may start by itself”