VidIQ - Video Intelligence

AI-powered video intelligence - download, analyze, clip, GIF from any URL. Supports YouTube, TikTok, Instagram, X. Uses ffmpeg + yt-dlp.

Security Scan

The skill's code and instructions are consistent with its stated purpose (downloading and manipulating videos using yt-dlp and ffmpeg), but there are a few minor omissions and privacy considerations to review before installing.

Purpose & Capability

Name/description match the included script: the shell script uses yt-dlp to download videos and ffmpeg/ffprobe to inspect, extract frames, clip, create GIFs, etc. The requested binaries (ffmpeg, yt-dlp) are the core tools needed. Minor mismatch: the script also invokes common utilities (python3, md5sum, bc, awk, grep, sed) that are not listed in the metadata but are typical on many systems.

Instruction Scope

SKILL.md tells the agent to 'feed frames to a vision model for content understanding' but does not specify which model or endpoint; the script itself only writes frames to /tmp/vidiq and does not transmit them. This leaves it to the agent runtime to decide where frames are sent, which could result in external transmission of image data if the agent forwards frames to a remote service. Also: SKILL.md lists a 'transcript' action in usage comments, but the script has no implemented 'transcript' case.

Install Mechanism

No install spec (instruction-only with an included shell script). This is low-risk from an installation perspective because no external code is fetched during install. Runtime does call yt-dlp which downloads video content (expected behaviour).

Credentials

The skill declares no required environment variables or credentials, which is appropriate for a local video manipulation tool. It does perform network downloads of user-supplied URLs via yt-dlp — expected for the feature, but note that any URL the agent is given will be fetched.

Persistence & Privilege

The skill does not request persistent/always-on privileges. It writes temporary files under /tmp/vidiq for caching and frames, which is normal for this functionality but means downloaded content is stored locally.

Assessment

This skill appears to do what it says: it downloads videos via yt-dlp and manipulates them with ffmpeg. Before installing, consider: 1) Privacy/exfiltration — the SKILL.md suggests sending extracted frames to a vision model but doesn't specify which; ensure you trust the agent or model that will receive those images. 2) Local storage — downloads and frames are cached under /tmp/vidiq; remove sensitive files after use. 3) Undeclared dependencies — the script uses python3, md5sum, bc, awk, grep, sed in addition to ffmpeg/yt-dlp; ensure your environment provides these. 4) Copyright and bandwidth — downloading videos may violate terms or consume network/disk. If you want tighter guarantees, request the author add explicit statements about where frame data is sent and list all required binaries in metadata.

VidIQ - Video Intelligence & Query Tool

Download, understand, and manipulate any video from a URL.

Commands

# Video info (duration, resolution, codecs)
{baseDir}/scripts/vidiq.sh <url_or_path> info

# Extract N frames evenly spaced
{baseDir}/scripts/vidiq.sh <url_or_path> frames 10

# Single frame at timestamp
{baseDir}/scripts/vidiq.sh <url_or_path> frame 01:30:00

# Clip between timestamps
{baseDir}/scripts/vidiq.sh <url_or_path> clip 01:01:01 01:20:01 output.mp4

# Create GIF (start time, duration in seconds)
{baseDir}/scripts/vidiq.sh <url_or_path> gif 00:45:00 5

# Extract audio as MP3
{baseDir}/scripts/vidiq.sh <url_or_path> audio

# Detect scene changes
{baseDir}/scripts/vidiq.sh <url_or_path> scenes 0.3

# Visual mosaic (columns, total frames)
{baseDir}/scripts/vidiq.sh <url_or_path> mosaic 4 16

AI Analysis Workflow

1. Extract frames: vidiq.sh <url> frames 10
2. Feed frames to vision model for content understanding
3. Answer questions about the video based on frame analysis

Supported Platforms

YouTube, TikTok, Instagram, X/Twitter, any direct video URL, local files.

Notes

• Downloaded videos are cached in /tmp/vidiq/ for reuse
• Frames output to /tmp/vidiq/frames_*/
• For long videos, extract more frames for better coverage
• GIFs are optimized with palette generation for small file size