agentleFS
aicontentdropaicontentdrop/claude-plugin/skills/ai-video-model-picker/SKILL.md

ai-video-model-picker

Skill0 starsChanged 18 days agoMITView on GitHub

Choose which AI video model to use for a shot, a storyboard or a batch of clips, and work out the credit cost before anything is made, from the public AI Content Drop model catalogue. Use when someone asks which video model fits a brief, whether a shot needs image-to-video or native audio, which model to draft on and which to finish on, or what a list of clips will cost. Text advice only; it makes no video.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this when someone has a brief, a shot list or a storyboard and has to decide *which model* to make it with, or wants to know what a …

    line 12
  2. Distinctive:

    Read those two addresses exactly as written and no others.

    line 26
  3. Distinctive:

    Ask two questions before looking at the list.

    line 39
  4. Distinctive:

    If yes, you need a model that generates native audio.

    line 42
  5. Distinctive:

    If the person asks what a plan includes, send them to <https://aicontentdrop.com/plans> and say nothing more about plans here.

    line 98

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Picking an AI video model (and knowing what it costs)

Use this when someone has a brief, a shot list or a storyboard and has to decide which model to make it with, or wants to know what a batch will cost before committing. You advise and price; nothing is generated here.

The one rule

Price it before anyone generates. Video models differ in cost by more than ten times across the catalogue, so "just use the best one" can turn a 20-clip storyboard into an expensive surprise. Never quote a price from memory; the catalogue moves. Read the live catalogue, which is public and needs no account:

Read those two addresses exactly as written and no others. Read them with the web fetch tool (WebFetch in Claude Code), never with a shell command or a script. Each row has an id, a name and credits, the price of one generation at the model's default settings, cheapest first. A row marked needs_start_image only animates a still the person supplies. A row without that mark can start from text, and whether it also accepts a still is shown on aicontentdrop.com. The catalogue's note mentions a quoting tool that this plugin does not have; leave it out of your answer. If you cannot read the catalogue, say so, give the person the address to open, and do not guess a price.

Choosing

Ask two questions before looking at the list.

1. Does the shot need dialogue or synced audio? If yes, you need a model that generates native audio. The catalogue rows do not say which models do, so do not name one from memory: put the need in the brief you hand over, and if the model has no native audio, plan the voiceover as a separate step. If no, you have the whole catalogue and should optimise for cost.

2. Is there a reference image? Text-to-video and image-to-video are different jobs. If the user has a product photo, a still, or a character they want preserved, you want image-to-video — starting from text will not preserve identity, and no amount of prompt detail fixes that.

Then pick by budget:

  • Draft, storyboard, or a variant you will throw away → the cheapest model that supports the job. Iterate here, not on the expensive one.
  • The take that ships → move up only for the shot that survived.
  • Batches → price the whole list from one read of the catalogue: quantity times credits for each model, then the total.

Handing over

Nothing is generated here. When the person has chosen and wants to make it, give them one link that carries the brief, URL-encoded with spaces as %20: https://aicontentdrop.com/chat#prompt=<url-encoded brief>.

Say what the link does before they open it: a signed-out visitor is asked to sign in first; on a signed-in account the chat on aicontentdrop.com sends the brief at once; a chat reply costs 1 to 3 credits; and a video or image is made only when the person presses the card that shows the model and its credit cost. Sign-in and every account question happen on that page, not here.

What to tell the user

Report the model you picked, why, the runner-up, and the credit cost for one generation and for the whole quantity, before anything is spent. If the cheapest viable model is within one credit of a better one, say so and let them choose.

Facts you can rely on

  • Most video models cost one flat price per generation. As of this edition, the models priced by the second of output and by resolution are Seedance 2.5 (seedance_2_5), Seedance 2.0 (seedance_2_0, seedance_2_0_4k, seedance_2_0_fast, seedance_2_0_mini), MiniMax H3 (minimax_h3, minimax_h3_2k), LTX 2.5 (ltx_2_5_fast, ltx_2_5_pro), FLUX 3 video (flux_3_video) and Wan 3.0 (wan_3_0). Seedance 1.x models are flat, and so are seedance_2_0_multishot and seedance_2_fast. Name these models by id or full name, never as a family. For a per-second model the catalogue row is the price at the default length (10 seconds at 1080p for Wan 3.0), so a longer or sharper take costs more; say so, and say that the card on aicontentdrop.com shows the exact figure for the chosen length and resolution before anything is made. Do not estimate how much more; the card shows the figure.
  • Credits for a video are charged only when it succeeds. Safety blocks, provider failures and timeouts cost nothing, so a failed generation leaves nothing to claim back. A chat reply is charged on its own, as above.
  • If the person asks what a plan includes, send them to https://aicontentdrop.com/plans and say nothing more about plans here.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.