02-url-to-explainer-video
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to fetch and read content from arbitrary external product URLs. This creates a surface for indirect prompt injection where an attacker could place malicious instructions on a webpage (e.g., in hidden HTML comments or metadata) designed to hijack the agent's behavior during the research and extraction phase.
- Ingestion points:
SKILL.mdGate 1 requires the agent to "fetch and read the actual page content before any other step". - Boundary markers: The instructions do not specify the use of delimiters or explicit system warnings to ignore embedded instructions within the fetched content.
- Capability inventory: The skill possesses the capability to generate scripts, select video providers, and compose final media files.
- Sanitization: No sanitization or filtering of the fetched content is described beyond the requirement to extract "claims".
- [COMMAND_EXECUTION]: The skill package includes executable scripts (
eval/run-eval.shandeval/contract_check.py) used for validating the skill's adherence to its defined gates. While these are intended for development and evaluation purposes, they constitute local executable code bundled with the skill.
Audit Metadata