Skills/ClawHub/video-sound

video-sound@1.0.9

CAUTION· static only

Publisher claims: "Make the sound of any video so it feels real: AI voices that don't sound fake (cast from candidates a person picks, directed with each model's documented tags), dialogue and narration written for the ear with pauses, breaths and small actions, sound effects on the right frame, music planned bar by bar with a quiet part, and a measured mix. Built from 480 sourced findings (Walter Murch, Randy Thom, Ben Burtt, trailer editors, NPR, voice directors, model docs, HyperFrames, earshot) plus paid tests. An agent can't hear, so every step writes numbers: a cue sheet, scripts that build and grade the mix (LUFS, true peak, voice over music, effects within a frame, cuts on the beat, dead air), a voice audition page and an audio LLM judge that shortlists takes. Use for "add sound", "voiceover", "sound design", "music for my video", "make the AI voice sound real", "pick the best take", "SFX", "mix", "launch video audio", "trailer sound", in HyperFrames, Remotion or AI films. Not for making a song as the product." Observed in sandbox: static analysis only — behavior not yet observed.

Registry ClawHubPublisher vargai · publishing for 215 daysInstalls 309Last detonated 16 h ago
What would fix thisWatchDispute verdict

Coverage

Static only; nothing was run yet.

The label rests on the code, the instruction files and the publisher. The honeypot run replaces this verdict when it completes.

Is [email protected] safe to install?

KillerSkills analysed it statically (a sandbox run is pending) on 2026-11-13 and scored it 60/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: HTML comment carrying imperative instructions to the agent. Summary: static analysis only — behavior not yet observed.

Behavior trace

Queued for honeypot run · static analysis only
t+CategoryObserved action

This version has not been detonated yet. The score above counts behavior as unknown (50/100), so it cannot read Verified until the honeypot run completes and replaces this verdict.

Static findings

MEDIUM
Python subprocess execution detectedscripts/judge_takes.py:103 · Python subprocess execution detectedout = subprocess.run(["ffprobe", "-v", "error", "-show_entries", "format=duration", "-of", "csv=p=0", path],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedscripts/audition.py:106 · Python subprocess execution detectedsubprocess.run(["curl", "-sL", "-A", "Mozilla/5.0", "-o", os.path.join(base, t["file"]), url], check=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedscripts/judge_takes.py:99 · Python subprocess execution detectedsubprocess.run(["ffmpeg", "-v", "error", "-y", *args], check=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
CRITICAL
HTML comment carrying imperative instructions to the agentREADME.md:1 · HTML comment carrying imperative instructions to the agentGenerated by .github/readme/build.ts from SKILL.md files and .github/readme/*.json. Edit those and run `bun .github/readme/build.ts`; CI fails when this file isFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedscripts/audition.py:116 · Python subprocess execution detecteddur = float(subprocess.run(["ffprobe", "-v", "error", "-show_entries", "format=duration", "-of", "csv=p=0", path],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectionreferences/evidence.md:389 · Prompt injection — known jailbreak technique reference · informational: not an instruction file…e the feeling that the only true resolution is to go out and watch the film." — Dan Graham (Owner of Gothic Storm trailer-music labels). https://www.soundonsound.c…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedscripts/sound.py:51 · Python subprocess execution detectedout = subprocess.run(["ffprobe", "-v", "error", "-show_entries", "format=duration", "-of", "csv=p=0", path],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedscripts/audition.py:114 · Python subprocess execution detectedreturn subprocess.run(["ffmpeg", "-hide_banner", "-nostats", "-i", path, "-af", af, "-f", "null", "-"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectionreferences/evidence.md:493 · Prompt injection — known jailbreak technique reference · informational: not an instruction file…egraded using an old, worn out tape effect." — Factory Studios sound designers (Dan Beckwith, Mark Hills, Phil Bolland) (London ad sound studio: Honda 'Hands', Joh…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
npx invocation of non-well-known packageSKILL.md:177 · npx invocation of non-well-known package — may execute untrusted codeautomation; `npx hyperframes audio <file> --json` reports a file's peak. **Remotion:** one `<Audio>` per stem;Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
LOW
Prompt injectionreferences/evidence.md:388 · Prompt injection — known jailbreak technique reference · informational: not an instruction file…tructure’: a compelling intro, a building main section and a massive ending." — Dan Graham (Sound On Sound) (Owner of three trailer-music labels under the Gothic S…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedscripts/sound.py:85 · Python subprocess execution detectedp = subprocess.run(["ffmpeg", "-v", "error", "-i", path, *(["-af", af] if af else []), "-ac", "1", "-ar", str(sr),Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectionreferences/evidence.md:573 · Prompt injection — known jailbreak technique reference · informational: not an instruction file…eus (Llama-based) takes 'name: text' prompts with voices tara, leah, jess, leo, dan, mia, zac, zoe and inline tags <laugh> <chuckle> <sigh> <cough> <sniffle> <groa…Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
MEDIUM
Python subprocess execution detectedscripts/sound.py:44 · Python subprocess execution detectedp = subprocess.run(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
LOW
Prompt injectionreferences/types.md:74 · Prompt injection — known jailbreak technique reference · informational: not an instruction file- Three acts, gaps between them as edit points, an ending that doesn't resolve (Dan Graham, Gothic Storm labels).Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
Prompt injectionreferences/music.md:80 · Prompt injection — known jailbreak technique reference · informational: not an instruction fileessential"; "Trailer endings typically reach a huge climax but don't resolve" (Dan Graham, Gothic Storm). MusicFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01
LOW
False authorization claimreferences/evidence.md:193 · False authorization claim — impersonating privileged role · informational: not an instruction file…s a consent recording from the same adult speaker reading a fixed statement. — "I am the owner of this voice and I consent to Google using this voice to > create a synthetic …Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01

Cost to the agent

Instructions loaded
≈3,398 tok
1 file · 24 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.

What would make this Verified

20 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim vargai: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  3. Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  4. npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
Publish a new version to ClawHub, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim vargaiPropose a patch to the maintainer