talking-head
Warn
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using
python docs/creative_cli.py. The--promptargument is populated with content derived from user-provided 'Message and goal' descriptions. This creates a risk for command or argument injection if the input contains shell metacharacters, as no sanitization step is defined. - [DYNAMIC_EXECUTION]: The skill performs dynamic runtime assembly of shell commands to interact with the FAL.ai pipeline via local Python scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a vulnerability surface for indirect prompt injection. 1. Ingestion points: User-supplied video messages, goals, and audience data are ingested during the intake phase. 2. Boundary markers: No delimiters or 'ignore' instructions are specified for the interpolated data. 3. Capability inventory: The skill has the capability to execute shell commands via a CLI tool. 4. Sanitization: No input validation or escaping is applied to the user-provided strings before they are incorporated into generated prompts and shell commands.
Audit Metadata