Skills

Verification

Use when checking whether agent edits are reflected in the OpenChatCut project and editor.

Install

npx skills add 0xsline/OpenChatCut --skill verification -g -a claude-code -y
Repository
0xsline/OpenChatCut2.8k
Category
AI and agents
Updated
14d ago
License
AGPL-3.0
Context cost
Lightabout 1,230 tokens
Runs scripts
No, instructions only

What it does

Use when checking whether agent edits are reflected in the OpenChatCut project and editor.

Verification

Use the lowest verification level that proves the requested result:

LevelRequired evidence
L0Static checks such as the focused verification script, npx tsc --noEmit, tests, and build.
L1A real Agent run against the editor at localhost:5199, followed by structural and rendered evidence.
L2The packaged desktop app completing the user scenario, including human visual review where automation is insufficient.

Runtime behavior changes require L0 + L1. Release and desktop-only changes also require L2 when the packaged app is the behavior under test.

Prefer two signals:

  1. read_project for structure: assets, tracks, items, frame placement, timeline duration.
  2. A visual capture path for rendered evidence at exact frames.

Use view_timeline_frames for composed timeline proof. This verifies the edited OpenChatCut timeline: trims, layers, captions, effects, markers, placeholders, crops, transitions, and layout.

For raw source-asset frame inspection, choose the cheapest path based on where the bytes live:

  • The agent in this build has no local filesystem access; all source bytes live in the project media store (/media/uploads/). Use view_asset_frames with the project asset id, the server takes an ffmpeg contact-sheet fast path automatically, so it is already the cheapest source-frame route.
  • view_timeline_frames renders the composed timeline (the editor-truth check); view_asset_frames samples raw source frames. Pick by what you are verifying.
  • There is no separate get_contact_sheet tool in this build, the contact sheet is what view_asset_frames / view_timeline_frames already return.

Use local/remote source-frame artifacts only for source understanding, moment selection, and rough trim decisions, not as edited output or timeline proof.

Read the whole SKILL.md on GitHub

More in AI and agents