What Is a Stickman Video Director? The Term Has Three Meanings
Stickman video director can mean a person, a GitHub project, or an AI tool that plans shots from a script. Here is what each one actually does.

If you search for a stickman video director, you get three unrelated results sharing one name: a job title, a code repository, and a browser tool that turns a short script into a shot list and a set of rendered clips. The confusion matters, because the three things do different work and are not interchangeable. This article separates them, then explains what the tool category actually does and where it stops.
The three things people mean by the phrase
A human director. In a small studio, the director is the person who decides how many shots a video needs, what each one shows, and in what order. They pick the wide shot that establishes the room, the close-up that lands the joke, and the cut that keeps a thirty-second clip from sagging. They also keep the stickman looking like the same stickman from shot to shot. That is a role, not a product.
A GitHub project. Plenty of searches for "stickman video director github" come from people hunting for a repository. A repository is source code: files, a README, open issues, a license. Someone has to install dependencies, supply API keys for a video model, and keep it running. It is not a hosted service with a login page, a credit balance, and a render queue. If you find one, read the README before assuming it is a finished tool.
A tool category. The third meaning is newer: a browser-based tool that reads a script and returns a directed shot list, then renders each shot with a video model. A stickman video maker of this kind sits between a text-to-video model and a blank page. You bring the script; the tool brings the plan. The result is a directed stickman video assembled from two to six separate shots, not one improvised prompt.
What a director adds to a stickman video
Stickmen are simple shapes, so most of the quality comes from decisions made before a single frame renders.
A shot list. How many shots, what each one shows, and in what order. A four-shot plan for a thirty-second clip might be: wide shot of an empty street, medium shot of the stickman walking in, close-up on the object he picks up, wide shot as he leaves. Written down, that list is easy to argue with. Improvised inside a single prompt, it is invisible until the render comes back wrong.
Camera grammar. A wide shot establishes. A close-up lands a point, and on a stickman a close-up of two dots and a curved line can carry a joke better than a full-body view. A slow push-in adds emphasis without a single word of dialogue. These are old film conventions, and they work on stick figures because the figure is not doing the acting — the framing is.
Continuity. The same line weight, head shape, palette, and character proportions across shots. This is the hard part, because each shot renders separately. A director checks that the stickman in shot three has the same head size as the stickman in shot one. Software can help — the first-frame workflow described below is the main lever — but continuity is something you enforce, not something you get for free.
Timing. A 5-second shot cannot carry a 40-word paragraph. Read a paragraph aloud at a normal pace and count the seconds; that is roughly how much room the shot needs, and if it does not fit, the paragraph becomes two shots or gets cut. This single check prevents most unusable renders.
Why script-first beats prompt roulette
A single long prompt asks a video model to invent story structure, camera placement, and continuity all at once, from one blob of text. The model has no reason to divide your idea into beats, so it usually produces one drifting scene. When the result is wrong, you cannot tell which part of the prompt failed.
A plan splits the script into units and sends one clear instruction per unit. Shot two is "stickman lifts the box, medium shot, slow push-in" and nothing else. If that shot misses, you rewrite one line instead of the whole prompt.
It is also cheaper. Planning a shot list costs nothing. You read the plan, decide whether it is worth rendering, and only then spend credits. On the default setting — 720p, 10 seconds — a shot costs 30 credits, so a four-shot video is 120 credits and you know that before you commit.
Your director's desk
Paste the script. Pick a stickman style. The director cuts it into shots and renders each one with Seedance 2.0. Planning is free — you only spend credits when you render.
1–400 words. One idea per shot works best.
Optional. Upload a stickman frame or paste a public image URL and the shot starts from it.
Credits are charged only when a shot starts rendering. A failed render is refunded automatically.
Director's plan
Paste at least one sentence to build the plan.
How the AI director builds the plan
Paste between 1 and 400 words. Punctuation and sentence boundaries become candidate shots: each sentence is a rough unit of action, which is why a script written in short sentences plans better than one long run-on.
The planner then merges short sentences and splits long ones until it has between two and six usable shots. A three-word sentence gets folded into its neighbor. A forty-word sentence gets broken where the action changes. Out of range, it keeps adjusting until the count fits.
Each shot receives an action line and a camera move, written plainly enough that you can edit them. From there you can generate one shot at a time: render the first, check the framing and the character, adjust the second, and only then spend credits on the rest. If you want help writing the action lines themselves, the stickman video prompt guide covers what those lines should and should not contain.
What you can control
Every shot is set independently, so a video can mix formats where that makes sense.
Style. Five options: Minimal (black on white), Midnight (white on near-black), Pop (red beanie, yellow tee), Prehistoric (charcoal on parchment), Cozy (warm daylight interior). Pick one and stay with it; switching mid-video breaks continuity faster than any camera mistake.
Format. Three aspect ratios (16:9, 9:16, 1:1), three durations (5, 10, or 15 seconds), and three resolutions (480p, 720p, 1080p). If you are publishing to more than one place, the guide to making stickman videos for YouTube covers which ratio to shoot first and how to reuse it.
A first frame. For image-to-video, upload a frame and the shot starts from it. This is how you keep a character matching an existing drawing rather than a description of one.
Cost visibility. The credit cost of each shot is shown before you render. The default is 720p at 10 seconds for 30 credits; 480p at 10 seconds is 15 credits and 1080p at 10 seconds is 60. Cost scales with duration. Credits are charged only when a shot starts rendering, and a failed render is refunded automatically. Plans, yearly options, and one-time packs are listed on the pricing page.
What it is not
It is not a frame-by-frame animation editor. There is no timeline, no keyframes, and nothing to install — it runs in a browser. If you want to hand-tune a fight scene frame by frame, that is a different kind of tool, and drawing skill still decides the result.
It is not a dialogue machine. Audio in the generated clip is limited to ambient sound effects that match the action. There is no voiceover, no dialogue, and no music track, so plan on adding those in an editor afterward.
It is not unmoderated. Text and image inputs are screened before a render starts, and if the moderation service is unavailable the request is refused rather than allowed through.
FAQ
Is there a real stickman video director GitHub project?
People use the phrase to search for code repositories, and repositories do exist for all sorts of video tooling. A repository is source code, not a hosted product — you run it yourself. This site is the hosted workflow: log in, paste a script, get a shot list, render shots.
Do I need to write a full screenplay?
No. 1 to 400 words is enough, and one sentence usually maps to one shot. Write the action in short, plain sentences and let the planner handle the grouping.
Does planning cost credits?
No. Credits are charged only when a shot starts rendering. You can build and revise a shot list without spending anything.
Can I use my own stickman drawing?
Yes. Upload a first frame and the shot starts from it, which is the most reliable way to keep a character consistent with an existing drawing.
Which format should I pick first?
720p at 10 seconds in 16:9 is the default for a reason: it balances cost and legibility. Ten seconds gives a sentence room to land, and 720p is sharp enough to judge whether the shot works before you pay for 1080p.