Asset Import Agent Skill
Use when acquiring or importing media into a OpenChatCut project asset library for video editing or creation, including local/attached videos, user-provided paths, public media URLs, web video/audio/image assets, upload fallback decisions, and deciding between import_media, download_media, or manual user action.
Install
one command, takes just this skill from the repositorynpx skills add https://github.com/0xsline/OpenChatCut --skill asset-importThe instruction itself
6 sections, as written by the authorAsset Import
This build runs the editor and its media server locally. There is no hosted import API, no CLI, and no OAuth: media enters the project through the editor UI or through the agent tools below. Pick the path by where the bytes live.
Path 1 — user's local files (primary)
Ask the user to drag the files into the editor (preview canvas or 我的素材 panel) or use the upload button. The local pipeline then runs automatically: streaming write to /media/uploads/, conditional transcode (≤1920 long edge, browser-friendly codec, ~8Mbps), audio extraction, and auto-transcription (ASR starts on upload). Do not ask the user to pre-convert, pre-trim, or transcode anything themselves — the pipeline handles it.
The agent cannot read the user's filesystem. If the user gives you a /Users/... or C:\... path, tell them to drop that file into the editor instead; you cannot fetch it.
Path 2 — public URLs (agent-driven)
Use download_media with up to 4 URLs per call. The server fetches, stores under /media/uploads/, and registers pool assets. Prefer this for stock/web media the user pointed at. After download, the asset behaves exactly like an upload (same transcode/ASR pipeline).
Path 3 — placeholder before bytes (timeline-first work)
Call import_media with {"action":"register_placeholder"} to mint a deterministic assetId and pool row before bytes exist. Use it when you want to lay out the timeline (edit_item etc.) while an upload is still in flight; the asset relinks automatically when bytes land. Report progress precisely: once the placeholder is registered say the assetId is known; while bytes are still uploading say so — never claim a file is ready before track_progress (target=upload) confirms it.
import_media with {"action":"create_session"} returns local direct-upload endpoints (POST /upload?name=...&assetId=...). This is for host-side scripts the *user* runs in their own terminal (e.g. curl -T file '/upload?...' against the dev server); the agent sandbox cannot reach localhost, so do not try to upload from run_code.
Editing discipline
For multi-source edits, build reviewable work from original source assets in the OpenChatCut timeline. Do not locally concatenate, pre-trim, pre-compose, burn captions, or flatten media before import — import originals and do all composition on the timeline so every step stays reviewable and undoable.
For code assets such as hand-authored Motion Graphics, use the create-motion-graphics skill (create_motion_graphic_from_code for new assets, asset-code updates for edits) — MG code is not an imported file.
Storage notes
- Bytes live on the local disk (
/media/uploads/), served directly with Range support. If Cloudflare R2 is configured, uploads mirror to R2 and other devices can hydrate from it; without R2 the project is local-only. - Local-only is a valid end state: preview, editing, and export all read from disk. Cloud mirroring is only needed for cross-device access.
Other skills for the same job
different authors, same section of the catalogue
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Improves the quality of images, especially screenshots, by enhancing resolution, sharpness, and clarity. Perfect for preparing images for presentations, documentation, or social media posts.
Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.
Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.
Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows.
Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications.
This skill should be used when working with pre-trained transformer models for natural language processing, computer vision, audio, or multimodal tasks. Use for text generation, classification, question answering, translation, summarization, image classification, object detection, speech recognition, and fine-tuning models on custom datasets.
How to use it
Take 0xsline/asset-import from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.