miniapi-video
Generate AI video, images and audio from Claude Code, Codex or any agent through MiniAPI, one prepaid key for every model. Seedance 2.5, Seedance 2.0, Seedance 2.0 Mini, Kling 3.0, Happy Horse, Wan 3.0 Prime and more. Text to video, image to video, reference to video, text to image, text to speech. Always quotes the price first (free) and asks before spending. Use when the user asks to make a video, animate an image, generate a clip, make a picture, make a voice-over, or asks for a cheaper Seedance, Wan or Kling API.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
01 前言 Front Matter
- name
- miniapi-video
- description
- Generate AI video, images and audio from Claude Code, Codex or any agent through MiniAPI, one prepaid key for every model. Seedance 2.5, Seedance 2.0, Seedance 2.0 Mini, Kling 3.0, Happy Horse, Wan 3.0 Prime and more. Text to video, image to video, reference to video, text to image, text to speech. Always quotes the price first (free) and asks before spending. Use when the user asks to make a video, animate an image, generate a clip, make a picture, make a voice-over, or asks for a cheaper Seedance, Wan or Kling API.
02 资源文件 1
- 脚本scripts/miniapi.mjs6.6 KB
03 SKILL.md 正文
MiniAPI video, image and audio
MiniAPI is one prepaid API key for video, image and audio models at https://api.minicut.ai.
A model marked coming soon in models has a price but takes no jobs yet; pick another.
Prices are per second or per image, set before the job runs. A job that fails is not charged.
Credits never expire.
Helper script (Node 18+, nothing to install): node scripts/miniapi.mjs <command>, run from this skill's folder.
Step 0. Key
- If
MINIAPI_KEYis not set, tell the user in one line: create a key at https://api.minicut.ai/console/keys?utm_source=skills-sh&utm_medium=skill&utm_campaign=miniapi-video , add funds ($10 minimum), thenexport MINIAPI_KEY=mapi_live_.... Wait for them. - Never print the key. Never write it into a file in the project.
Step 1. Pick the model
Run node scripts/miniapi.mjs models for every id with its starting price.
Run node scripts/miniapi.mjs model <id> for that model's parameters, limits and prices.
Use only parameters listed on that model's page. A parameter the model does not take is refused with 422.
Defaults when the user does not name one:
| Job | Model id | Why |
|---|---|---|
| Video, best quality | seedance-2.5 |
4 to 30 s, up to 1080p, references, sound on by default |
| Video, cheap draft | seedance-2.0-mini |
from $0.054 / s |
| Image | read models and pick the image model the user names, else the cheapest image model |
Step 2. Quote, then ask
- Build the JSON body from the model page.
- Run
node scripts/miniapi.mjs quote <id> '<json>'. A quote is free and runs nothing. - Show the user the PRICE line and ask for a yes. Never submit without it.
- One yes covers one job. A retry or a second clip is a new quote and a new yes.
Step 3. Run
node scripts/miniapi.mjs run <id> '<json>' --yes --out <folder>
It quotes again, submits, waits, downloads the file into <folder> and prints SAVED <path> and CHARGED <price>.
Show the user the saved path. For a picture, open it. For a clip, give the path.
Errors
| Exit | Meaning | Do |
|---|---|---|
| 3 | no key | Step 0 |
| 4 | balance too low | tell the user the price and the add-funds link it printed |
| 5 | MiniAPI refused the body | read the message, fix the field from the model page, quote again |
| 7 | the job failed | not charged. Say so, and offer one changed retry with a new quote |
Example
User: "make a 5 second clip of a red fox running through snow at dawn, 720p"
node scripts/miniapi.mjs quote seedance-2.5 '{"prompt":"a red fox runs through snow at dawn","resolution":"720p","duration":5}'
# PRICE 1.750000 USD ... Quote is free. Nothing ran.
# (ask the user, wait for yes)
node scripts/miniapi.mjs run seedance-2.5 '{"prompt":"a red fox runs through snow at dawn","resolution":"720p","duration":5}' --yes --out clips
Docs: https://api.minicut.ai/docs . Every model as Markdown: https://api.minicut.ai/llms.txt
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.