ttdl

Warn

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/setup script modifies shell configuration files (e.g., .bashrc, .zshrc, .profile) to append ~/.local/bin to the PATH environment variable. This persistence mechanism is used to install the CLI tool but involves automated modification of sensitive user environment files.\n- [COMMAND_EXECUTION]: The scripts/ttdl tool executes the ffmpeg binary with arguments including durations and output filenames that are derived from TikTok page metadata.\n- [REMOTE_CODE_EXECUTION]: The skill uses browser-harness-js to execute a JavaScript heredoc that interacts with the filesystem via the node:fs module and performs browser automation via the Chrome DevTools Protocol. This dynamic execution of logic is the primary engine of the skill.\n- [PROMPT_INJECTION]: The skill processes untrusted external content, specifically TikTok video titles and description metadata. Although character-based sanitization is applied before using this data in shell commands, it remains a surface for indirect injection if malicious metadata is encountered on a page.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 17, 2026, 04:46 PM
Security Audit — agent-trust-hub — ttdl