Youtube Editor
ReviewAudited by ClawScan on May 1, 2026.
Overview
The skill’s disclosed behavior matches its YouTube editing purpose, but it uses external AI services, API keys, local subprocesses, and optional helper-skill execution that users should understand before use.
Before installing, confirm you are comfortable sending the target video/audio and transcript content to OpenAI, use API keys with appropriate limits, install FFmpeg and Python dependencies from trusted sources, and review nano-banana-pro separately if you plan to use AI thumbnail generation.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Private or unpublished video/audio content may be processed by OpenAI.
The script sends audio and transcript-derived prompts to OpenAI for transcription and metadata generation. This is central to the skill’s purpose, but it means user video content leaves the local machine.
client.audio.transcriptions.create(model="whisper-1", file=f, response_format="srt") ... client.chat.completions.create(model="gpt-4o", messages=[{"role": "user", "content": prompt}])Use the skill only for content you are comfortable sending to OpenAI, and review your OpenAI account’s privacy, retention, and billing settings.
The optional image-generation key may be used by the helper skill and could incur charges or be exposed if the helper or local environment is not trusted.
The optional image-generation API key is read from the environment and forwarded to the nano-banana-pro helper command. This is purpose-aligned for thumbnail generation, but it entrusts the credential to another installed skill process.
api_key = os.getenv("NANO_BANANA_KEY"); if api_key: cmd.extend(["--api-key", api_key])Set NANO_BANANA_KEY only when using thumbnail generation, use limited-scope or low-quota keys where possible, and review nano-banana-pro separately before enabling that feature.
Running the skill will execute local video-processing tools and may execute the optional helper skill if installed.
The script runs FFmpeg and optionally runs a fixed-path helper script through uv. Local command execution is expected for video processing and thumbnail generation, and the code uses argument lists and timeouts.
subprocess.run(cmd, check=True, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL, timeout=900) ... cmd = ["uv", "run", skill_path, "--prompt", prompt, "--filename", output_file, "--resolution", "4K"]
Install FFmpeg and any helper skill from trusted sources, and avoid running the optional thumbnail path unless you have reviewed the helper skill.
Future package updates could change behavior or introduce dependency risk.
The uv script dependencies are not version-pinned. These packages are expected for the stated workflow, but unpinned dependencies can change over time.
# dependencies = ["openai", "yt-dlp", "playwright", "rembg[cpu]", "pillow"]
Prefer pinned dependency versions or a lockfile if using this skill in a repeatable or production workflow.