agent-media-v2
Pass
Audited by Gen Agent Trust Hub on May 27, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
agent-mediaCLI commands to list characters, check job status, and generate video assets. These operations are essential to the skill's primary function and are performed following a structured user-confirmation workflow. - [EXTERNAL_DOWNLOADS]: The README provides instructions for installing the
agent-media-cliand@agentmedia/mcp-serverpackages from the standard NPM registry. These packages are official components provided by the skill author. - [PROMPT_INJECTION]: The skill includes instructions to guide agent behavior (e.g., 'HARD GATE', 'NEVER mention pricing'). These are design constraints intended to ensure high-quality output and maintain a specific user experience, rather than attempts to bypass safety filters or override system-level instructions.
- [DATA_EXPOSURE]: The skill handles
character_idand URL references for media assets. It specifically instructs the agent to handle these IDs internally and only present human-friendly names to the user, preventing unnecessary exposure of internal identifiers. - [SAFE]: No obfuscation, persistence mechanisms, or privilege escalation patterns were detected in the analyzed files.
Audit Metadata