good-TTvideo2text

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (HIGH): The skill instructs the agent to execute shell commands using raw user input as arguments (e.g., python scripts/transcribe.py "USER_PROVIDED_TEXT"). This presents a critical command injection vulnerability if the agent or script fails to sanitize the input against shell metacharacters.
  • [PROMPT_INJECTION] (HIGH): This skill is vulnerable to Indirect Prompt Injection. Ingestion points: Raw user share text and video metadata (title, author). Boundary markers: None defined. Capability inventory: Subprocess execution and network access. Sanitization: Recommended in text but the execution examples show raw string interpolation. Attackers could embed malicious instructions in video descriptions to hijack the agent.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill requires installing unverified dependencies via requirements.txt and includes an unvetted local library (TikTokDownloader/), posing a risk of supply chain attacks.
  • [CREDENTIALS_UNSAFE] (LOW): The skill manages session cookies in TikTokDownloader/settings.json, which could lead to credential exposure if the local environment is accessed by an attacker.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 08:23 AM