← Back to Skills Marketplace
cawcut

VN Skill

by cawcut · GitHub ↗ · v0.1.0 · MIT-0
darwin ⚠ suspicious
115
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install vn-skill
Description
Local video, audio and image processing expert for macOS, powered by VN Video Editor. Use this skill whenever the user wants to process video, audio or image...
Usage Guidance
This skill appears to do what it says (local VN-based media exports), but it installs a helper CLI by downloading and extracting a zip from a third-party GitHub release. Before installing: 1) Inspect the GitHub release and repository (owner: cawcut) to confirm it is legitimate and maintained by a trusted author. 2) Verify the downloaded file's SHA256 matches the value in metadata. 3) Optionally open the zip and review the binary contents or run the CLI in a sandbox/container first. 4) Be aware auto-caption engines may download ML models on first use (network activity). If you are not comfortable executing an unsigned binary from that release, decline installation and use alternative tools (ffmpeg or official VN tooling).
Capability Analysis
Type: OpenClaw Skill Name: vn-skill Version: 0.1.0 The skill facilitates local media processing by downloading and executing a third-party binary (vnapp-cli) from GitHub. While the installation logic in SKILL.md includes a SHA-256 checksum verification (857417f671fd6f38ec30cf993f80c79efede7641f3532c2e9a99e75177a5853f), it performs high-risk actions such as programmatically bypassing macOS Gatekeeper (xattr -d com.apple.quarantine) and executing downloaded code. These behaviors are consistent with the stated purpose of bridging to the VN Video Editor GUI, but the reliance on an external binary and the suppression of OS-level security warnings constitute a significant supply-chain risk.
Capability Tags
cryptorequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The name/description (local VN Video Editor bridge for macOS) matches the instructions and CLI reference: operations are local, target macOS, and require the VN app. However the registry summary claims 'No install spec / instruction-only', while the SKILL.md metadata includes a download/install entry for a vnapp-cli binary — this inconsistency should be clarified.
Instruction Scope
SKILL.md limits actions to local file processing (video/image/audio paths supplied by the user) and session-scoped preferences. It does mention that some auto-caption engines ('whisper_turbo', 'whisper_medium') may download models on first use — network activity that is outside pure local file I/O but still consistent with running local ML models. The instructions do not request unrelated files, environment variables, or secret exfiltration.
Install Mechanism
The skill metadata contains an install entry that downloads and extracts a zip from a GitHub Releases URL (https://github.com/cawcut/skill-vn/releases/...). Extract=true means code will be written to disk (~/.openclaw/tools/vnapp-cli/). The release URL and a sha256 are provided (good), but because the top-level registry summary claimed 'No install spec', this installer is effectively hidden unless the user inspects SKILL.md. Any downloaded executable carries risk; although GitHub Releases is a normal distribution host, archive extraction and running an unsigned binary should be treated cautiously.
Credentials
The skill does not request environment variables, credentials, or config paths. It requires the VN Video Editor app and that VN's MCP server be active, which is proportional to the stated purpose.
Persistence & Privilege
always is false and the skill does not request system-wide privileges or modify other skills. The installer writes into the user's home under ~/.openclaw/tools which is reasonable for a user-scoped tool, but it will add an executable to the local system.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install vn-skill
  3. After installation, invoke the skill by name or use /vn-skill
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
Initial release — auto-captions, burn SRT, compress, denoise, extract, merge, background removal; local processing on macOS via VN Video Editor
Metadata
Slug vn-skill
Version 0.1.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is VN Skill?

Local video, audio and image processing expert for macOS, powered by VN Video Editor. Use this skill whenever the user wants to process video, audio or image... It is an AI Agent Skill for Claude Code / OpenClaw, with 115 downloads so far.

How do I install VN Skill?

Run "/install vn-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is VN Skill free?

Yes, VN Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does VN Skill support?

VN Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin).

Who created VN Skill?

It is built and maintained by cawcut (@cawcut); the current version is v0.1.0.

💬 Comments