Talking Head Prompter
Reading a script makes you sound robotic. Winging it costs you twelve takes. Prompt cards sit between the two — bullets to trigger the thought, written openings and transitions to stop the stumbles, energy notes so card 5 doesn't sound flatter than card 1.
What this skill does
There are two ways a talking-head video falls apart. The first is reading a script — the eye-to-mouth lag is detectable within ten seconds and viewers tune out. The second is winging it — rambling, missing points, doing twelve takes and still feeling like the message didn't land. Prompt cards solve both. They give you the structure that confident delivery fills in. News anchors, keynote speakers, and professional YouTubers all use this format because it works.
Each card is a self-contained 1-3 minute block with a fixed shape. A written opening line to get you started cleanly — no "uhh, so..." moment. Three to five bullet points that trigger the thought without scripting the sentence. A written transition line that bridges to the next card so you never have to improvise the awkward bit. An energy note that tells you where to lean in, where to slow down, where to add emphasis — because without those, presenters default to one tone for the whole video and the energy drops off after card three. The single most important word in each bullet is bolded — that's what your eye catches in a glance before looking back at camera.
The card count scales to your runtime — roughly one card per 1.5-2 minutes — and the design assumes one card equals one take. Nail cards 1-4, fumble card 5, and you only reshoot card 5. For screen-share work the cards become stage direction sheets too, with [SHOW: click on X] action markers inline so you're not scrambling between tabs while recording. Ad-lib zones get marked where your own stories should go — the prompt tells you what kind of story, you supply the actual one, which is where the personality lives.
For nervous or first-time presenters the skill leans heavier on opening and transition lines and lighter on bullets per card — more safety rails, shorter takes, confidence building with repetition. If you really want word-for-word, you get a teleprompter version as a companion, with a note that cards usually produce more natural delivery and to try them first. The whole point is to look prepared instead of mid-thought, without the cost of sounding read.
When this triggers
- ·You know your topic cold but freeze the moment the red light comes on
- ·Your last video took 14 takes and you can't face doing that again
- ·You're doing a screen-share walkthrough and need talking points plus action cues on one card
- ·You hate the teleprompter sound but a pure outline leaves you mid-sentence
- ·First time on camera and you need more guardrails than an experienced presenter would
Example
Trigger
User: '10-minute video on positioning for freelance designers. Sitting, talking-head, slightly nervous on camera.'
Output
5 cards × ~2 min each. Tone: direct, conversational. CARD 2 OF 5: THE TWO-AXIS TRAP (2:00-4:00) OPEN: "Most positioning advice uses two axes. I think they're both wrong, and here's why." · **Axes** — what frameworks tell you to pick [PAUSE] · **Problem** — they assume you have a sales team · [YOUR EXAMPLE: a time you tried a 2-axis framework and it didn't work for you] · **Reframe** — what to use instead TRANSITION: "OK so if not two axes, what? Let me show you the single-vector model." ENERGY: Lean in for the "I think they're wrong" line. Slow down on the example. Eye contact back to camera before the transition. + filming checklist, time budget, optional teleprompter version.
Get this skill + 7 more
Included in the The Creator Stack — create more, better, faster. Save $120+ vs buying individually.
Get The Creator Stack — $99What you get
- 147-line SKILL.md, ready to drop into ~/.claude/skills/
- Tested through 3 Karpathy-loop iterations (versions v1.0.0 → v1.3.0)
- Triggers automatically when relevant — no command to remember
- Lifetime updates as the skill is refined further
More from Video Production
Takes raw material — interviews, research, topics, and footage descriptions — and structure them into a 3-act narrative arc that turns information into a story people can't stop watching
Takes a video topic and generate 5 distinct thumbnail concepts — each with specific text, emotion, composition, and contrast direction
Takes a topic and generate 20 opening hooks — each tested against scroll-stopping criteria and categorised by psychological mechanism
Writes scripts designed for maximum audience retention — not just good information, but information delivered in a structure that keeps people watching
Takes a video's title, description, and topic and produce fully optimised metadata — title, description with timestamps, tags, and chapter markers
Takes long-form videos, ideas, or topics and distil them into 3-5 Shorts scripts under 60 seconds each
Browse the full library
297 skills across 31 categories. One purchase, lifetime updates.
See all bundles