Communitygithub.com

long-video-story

Agree on the story for a long video. The user either supplies a story or asks Claude to write one. When Claude writes, it first asks which story style to use (for example niche discovery, or a provocative-questions narration), runs that style's skill, and revises until the user approves. Use after long-video-setup, or when the user says "story for my video", "give me an idea", or wants to change an unapproved story. Ends only when the user explicitly approves the story.

What is long-video-story?

long-video-story is a Claude Code agent skill that agree on the story for a long video. The user either supplies a story or asks Claude to write one. When Claude writes, it first asks which story style to use (for example niche discovery, or a provocative-questions narration), runs that style's skill, and revises until the user approves. Use after long-video-setup, or when the user says "story for my video", "give me an idea", or wants to change an unapproved story. Ends only when the user explicitly approves the story.

Works with✓Claude Code~Codex CLI~Cursor
npx skills add https://github.com/samanfirouzi/long-video-skill/tree/HEAD/skills/long-video-story

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentation

Long video story

Stage 2 of the long video workflow. Output: an approved story.md in the project folder.

This stage ends only when Claude and the user agree on the story. No later stage starts before that.

Before you start

  1. Find the active project: ~/.long-video/projects/<slug>/project.json. If the user has several, ask which one.
  2. If no project exists, run long-video-setup first.
  3. Load the project's config from ~/.long-video/configs/<config-name>.json. Use it for length, language, audience, tone, format and visual approach. Do not re-ask what the config answers.
  4. If story.md already exists with status draft, offer to continue it. If status is approved, ask whether to reopen it.

Step 1: Ask the entry question

Ask: "Do you have a story, or do you want me to write one?"

  • I have a story. The user pastes or describes it. Go to Step 5.
  • Write one for me. Go to Step 2.
  • I have a rough idea. The user gives a seed (topic, theme, a line). Keep it, then go to Step 2.

Step 2: Ask for the story style

Always ask this when Claude writes the story. Do not skip it, and do not pick a style for the user.

  1. Read references/story-styles.md. Check which style skills are installed.
  2. Show the available styles, each with its one-line description. Always include open.
  3. Ask the user to pick one. They can also describe their own style. If they do, treat it as open with their description as extra instructions.
  4. Save the choice in project.json as "story_style": "<id>".

Step 3: Get the topic

Follow the chosen style's type.

  • ideas (for example niche): call the style skill. Pass the config summary and any seed. It returns a topic. Then ask which writer to use (Step 4 pairing rule in the registry).
  • writer (for example questions): if the user has a topic or seed, use it. If not, ask for one. Do not suggest topics.
  • open: pitch ideas, as below.

Open style: pitch ideas

  1. If there is no seed, ask at most 3 short questions: topic area or interests, something to avoid, any real events or sources to use.
  2. Write 4 pitches that fit the config. Each pitch has: title, one-sentence logline, the first 15 seconds as a hook, and why it works for this audience.
  3. Make the 4 pitches clearly different from each other.
  4. Ask the user to pick one, combine two, ask for more, or give a new direction. Repeat until the user picks.

Step 4: Write the story

Call the writer.

  • A writer skill (for example long-video-story-questions): call it with the topic, the target duration, the language and the config summary. Its own output format wins. Do not reshape it into the default structure.
  • open: write the default structure below.

Default structure (open style)

Match the target length in the config. Spoken narration runs about 150 words per minute, so a 15 minute video needs about 2,250 spoken words. Do not write the final script now. Write the story: what happens, in order, with enough detail for the user to judge it.

  1. Title (working).
  2. Logline.
  3. Hook. The opening 15 to 30 seconds.
  4. Story outline. Acts or chapters. For each: what happens, the turning point, rough share of runtime.
  5. Full story. Prose retelling, beginning to end.
  6. Ending and call to action.
  7. Facts to verify. List every factual claim that needs a source. Write "none" for pure fiction.

Rules for the open style:

  • Do not invent facts for a non-fiction story. Mark unknowns in "Facts to verify".
  • Do not copy existing works. Use original structure and wording.
  • Keep language and tone from the config.

Save the draft

Save to story.md with this header, status draft:

---
title: <working title>
status: draft
revision: 1
style: <style id>
writer: <writer id or "open">
config: <config-name>
---

Put the text below the header. Show the story in chat.

Step 5: Confirm or revise

If the user brought their own story in Step 1, save it first as story.md with the header above (style: user).

Ask: approve, or request changes. Ask in a new message after the story. Do not put the question inside a writer skill's output.

  • Changes requested. Apply them. If a writer skill made the text, keep its rules (format, tone, honesty) while editing. Increase revision. Show what changed in two or three lines, then show the updated story. Ask again.
  • Question or doubt. Answer it. Offer a change if the answer shows a problem.
  • Wants a different story. Go back to Step 2. Keep the old file as story.rev<N>.md.
  • Approve. Needs an explicit yes. Silence, "looks fine so far" or "ok but..." is not approval. If unclear, ask one direct question: "Lock this story?"

Loop until approved. Do not rush the user.

Step 6: Lock

  1. Set status: approved in story.md front matter. Add approved: <ISO date>.
  2. Update project.json: set "stage": "story-approved".
  3. Tell the user the file path. Say the story stage is done.
  4. Offer to continue with long-video-assets (characters, environments, objects and style). Start it only if the user agrees.

Rules

  • Never mark a story approved without an explicit yes.
  • Never change an approved story silently. Reopening needs the user's request, and sets status back to draft.
  • Never skip the style question when Claude writes the story.
  • Never call a style skill that is not installed.
  • Keep pitches and revisions short. Show full text only for the story itself.

Individual skills in this repo

This repo contains 6 individual skills — each has its own dedicated page.

long-video-assets

Video and animation skill. Build the visual foundation for a long video from an approved story. Extracts the recurring characters, environments and key objects, helps the user choose one art style from suggested options (10 at a time, more on request), then writes image generation prompts for every character (portrait plus character sheet), environment and object, all sharing one locked style. Can also generate the images (installed tool, API, MCP or browser) or hand the user a folder and file names to fill. Use after lon

long-video-director

Turn an approved story into a shot-by-shot video script, acting as the director. Asks whether the video uses a narrator or characters who speak (or both), suggests professional filming choices (camera type, lenses, camera movement, lighting, color, pacing), saves them as a separate film style file, then writes a script that matches the project config. Outputs film-style.md and video-script.md. Use after long-video-story (and long-video-assets) are done, or when the user asks for the video script, shot list, camera style or lighting plan.

long-video-keyframes

Create the start and end frame image for every scene (clip) of the approved video script. Writes one image prompt per frame with a fixed file name that holds the scene number and start or end, makes a folder for the images, and can generate the images with an available image tool, or hand the prompts to the user all at once or one by one. Use after long-video-director, or when the user asks for start frames, end frames, keyframes or scene images.

long-video-render

Assemble the finished video and run the review loop. Builds the edit from the clips (or images with camera moves), voice, sound stems and captions, renders a low quality review version with scene numbers and time burned in so a client can review it, collects and applies changes (re-rendering reviews as often as needed), and when the user approves, renders the high quality final video. Uses a video editor with a CLI or MCP (FilmCraft, EffectCraft or another) if one is available and the user wants it, otherwise the bundled ffmpeg renderer. Use after long-video-sound, or when the user asks to render, assemble, export, preview or finalize the video.

long-video-sound

Video and animation skill. Create the sound effects, ambience and background music for the video and place them on the timeline. Plans the sound design from the approved script, sources or makes each sound (Openverse for openly licensed recordings and music, Earshot to make sounds from words and to listen to and judge sounds, Clipwing real-sfx to place real sounds on exact frames), records licences and credits, and mixes separate stems (effects, ambience, music) with the music ducked under speech. Or gives the user a fold

long-video-voice

Create the spoken audio (narration and character voices) for the approved video script. First decides whether the video needs speech and asks the user to confirm. If yes, builds a voice cast (narrator plus each speaking character), makes a short sample of each voice with an available tool such as an ElevenLabs MCP and asks the user to approve the voices, then generates one audio file per spoken line with a name that holds the scene number and speaker. Or gives the user the lines, file names and folder to record or generate themselves. Use after long-video-clips (or long-video-keyframes), or when the user asks for narration, voiceover, voices or dialogue audio.

Related Skills