Writes documentary narration from picture descriptions and research notes. Drafts commentary that adds context without describing what is already visible on screen.
How this skill is triggered — by the user, by Claude, or both
Slash command
/autopunk-media-skills:commentary-writerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Writes narration or commentary for a documentary sequence, given a description of the picture and research notes or factual context.
Writes narration or commentary for a documentary sequence, given a description of the picture and research notes or factual context.
Required: A description of the sequence: what is seen on screen, shot by shot or section by section; the factual context or research notes the narration should draw on; the approximate total narration length needed (in words or seconds) Optional: The film's narrator voice profile (formal/conversational, detached/intimate, authoritative/questioning); the tone of the surrounding sequences (for tonal consistency); any specific facts or quotes that must appear in this section; the film title and general subject
Formatted commentary script. Each narration block labeled with approximate screen position (e.g., "Over the departure shot," "Into the archive footage"). Word count per block noted. Narration text in regular weight; picture notes in italics. Total word count at the end. Separate Commentary Notes section flagging any sections where the picture description suggests narration should be held or where the factual content was thin.
Sequence: A documentary about the last season of a traditional nomadic herding community before they settle permanently. Picture sequence:
Research notes: This community has herded this route for at least four generations. The government's sedentarization program has offered land and housing subsidies to nomadic families, accelerating settlement. This season, Altansarnai's family will move to a permanent house for the first time. She estimates the route they are covering today has been walked by her family for over 100 years. The family has 340 sheep and 28 horses. The journey between summer and winter pastures takes approximately three weeks. Narration length: Approximately 120–140 words total Voice: Quiet, measured, authoritative. No sentimentality. The narration should carry weight without editorial emotion.
Commentary Script — Herding Community Opening Sequence Approx. total narration: 128 words
[Wide landscape shot — sunrise, yurts in middle distance]
Narration (over landscape, before Altansarnai appears): "For four generations, Altansarnai's family has herded this steppe. Sheep in summer, horses year-round, the same route between the same pastures. The journey between them takes three weeks." (~35 words)
[Altansarnai emerges. Morning tasks. Children loading. Older family members directing.]
Narration (over the loading sequence): "This year, a government subsidy has made it possible — and, for many, inevitable — to stop. A permanent house. Land rights. An end to the route." (~30 words)
[Hold here. Let the image work.]
[The herd being gathered.]
Narration (brief, as herd comes together): "Three hundred and forty sheep. Twenty-eight horses. A last season." (~10 words)
[Pause. Let the numbers settle.]
[Caravan moving away from the empty site.]
Narration (over the departure shot — this is the sequence's emotional weight): "Altansarnai will tell you she is not sad. She says it practically: the route is harder each decade, the winters longer, the children want something different. She does not say what she will miss." (~38 words)
[Hold on the landscape after the caravan clears frame. Silence.]
Commentary Notes
npx claudepluginhub ur-grue/autopunk-media-skills --plugin autopunk-media-skillsTransforms research notes, transcripts, or articles into broadcast-ready voice-over scripts with short sentences, active voice, and timing estimates for picture.
Writes and validates timestamped Chinese narration scripts for analyzed videos. Use after video-understanding produces analysis files. Outputs validated narration.json.
Transforms user storytelling prompts into structured video scripts (JSON) with visual cue tags, voice selection, and orientation settings for a video generator tool.