fal-ai-media
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
fal-ai-mcp-serverpackage usingnpxto provide the necessary tools for media generation.\n- [DATA_EXFILTRATION]: Performs legitimate network requests tofal.aiandapi.elevenlabs.ioto transmit prompts and receive generated media content.\n- [COMMAND_EXECUTION]: Executes shell commands vianpxto initialize the MCP server and uses Python code to write generated audio data to a local file (output.mp3).\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing user-supplied prompts for media generation.\n - Ingestion points: User-provided
promptandtextvalues in MCP tool calls and Python snippets (SKILL.md).\n - Boundary markers: The skill utilizes structured JSON payloads and function parameters to encapsulate user input.\n
- Capability inventory: Network communication with API endpoints, local file system writes, and shell command execution.\n
- Sanitization: No explicit prompt validation or sanitization logic is present in the examples.
Audit Metadata