Skillv1.0.2

ClawScan security

ffmpeg-static · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 19, 2026, 6:39 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's files, requirements, and instructions match its stated purpose (providing a Node-friendly FFmpeg binary and usage patterns); nothing requested is disproportionate or unrelated to FFmpeg usage.
Guidance: This skill is coherent for bundling and using an FFmpeg binary in Node projects. Before installing: (1) confirm you trust the npm package and its GitHub releases (ffmpeg-static downloads pre-built binaries during npm install); (2) if you operate in a locked environment, consider using system FFmpeg and set FFMPEG_STATIC_SKIP_BINARY_DOWNLOAD=1; (3) never run FFmpeg on untrusted filenames or URLs without validation—FFmpeg can read network resources and many protocols; (4) prefer system FFmpeg for hardware-accelerated encoders and to avoid unexpected binary downloads; (5) verify the package on npmjs/github and inspect hashes or CI reproducible builds if you need high assurance.

Review Dimensions

Purpose & Capability: noteName/description, required binaries (node, npm), templates, and the resolve script all align with packaging and exposing ffmpeg-static to Node projects. Minor note: the SKILL.md language says it "wires ... into every assistant interaction," which suggests broad integration, but the skill metadata shows always:false (it is user-invocable). This is a wording mismatch rather than a functional inconsistency.
Instruction Scope: noteSKILL.md contains explicit instructions for resolving and invoking FFmpeg from Node and shell; it recommends spawning child_process and provides many FFmpeg command templates. These instructions stay on-topic for media processing. Two security-relevant points are called out by the skill itself: (1) npm install of ffmpeg-static causes pre-built binaries to be downloaded from GitHub Releases, and (2) FFmpeg can read from URLs/protocols so running it against untrusted inputs may reach network resources or other protocols. Those are expected for this capability but worth awareness.
Install Mechanism: noteThere is no platform-specific installer in the manifest, but SKILL.md recommends npm install ffmpeg-static (and optionally ffprobe-static). Using npm is expected; ffmpeg-static performs post-install binary retrieval from GitHub Releases (pre-built FFmpeg). This is moderate-risk in general (network download of binaries) but is the standard distribution method for this package and not an unexpected arbitrary URL or custom server.
Credentials: okThe skill declares no secrets and only documents benign, proportional environment variables (FFMPEG_PATH, FFPROBE_PATH, FFMPEG_STATIC_SKIP_BINARY_DOWNLOAD). No unrelated credentials or config paths are requested.
Persistence & Privilege: okalways is false, and the skill does not request permanent/invisible presence or system-wide config modifications. disable-model-invocation is false (normal); the skill can be invoked by agents but that is expected for user-invocable skills.