youtube-shorts-generator
Warn
Audited by Gen Agent Trust Hub on Jun 25, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to clone an external repository from GitHub (
https://github.com/SamurAIGPT/AI-Youtube-Shorts-Generator.git) and execute its scripts (python main.py). This allows for the execution of arbitrary remote code that is not part of the skill's own distribution. - [EXTERNAL_DOWNLOADS]: The pipeline requires installing third-party dependencies using
pip install -r requirements.txt. These dependencies are not version-locked or verified, creating a supply chain risk if the requirements file or the registry is compromised. - [COMMAND_EXECUTION]: The skill uses shell commands and subprocesses (
python main.py,xargs) to perform video downloading, transcription, and processing. While intended for its primary purpose, this execution environment could be abused if user input (URLs or arguments) is not strictly sanitized. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8).
- Ingestion points: Processes untrusted data from YouTube video content, specifically the transcript generated by the Whisper API.
- Boundary markers: No explicit boundary markers or instructions to ignore embedded commands within the transcript are provided.
- Capability inventory: The skill possesses the capability to execute shell commands, read/write local files, and make network requests to the MuAPI.
- Sanitization: There is no evidence of sanitization or filtering of the transcript text before it is passed to the LLM for highlight ranking and classification, which could allow malicious instructions embedded in a video to influence the agent's behavior.
Audit Metadata