mage-generate

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an extensive instructional framework for an AI agent to interact with the Mage generative AI platform through an MCP (Model Context Protocol) connector. It contains detailed documentation for various image, video, and audio models (e.g., Mango, Cherry, Guava, ElevenLabs).
  • [SAFE]: The workflow for handling local file uploads via the create_upload tool and subsequent curl or ffmpeg commands is standard practice for media processing agents and is well-documented within the skill's reference materials.
  • [SAFE]: All external domains and resources mentioned, such as mage.space and mcp.mage.space, are official vendor-owned infrastructures. There are no signs of credential exfiltration, malicious downloads, or unauthorized access to sensitive files.
  • [SAFE]: The skill explicitly instructs the agent to respect content moderation policies and not attempt to bypass safety filters (e.g., OpenAI's filter for GPT models or Mage's own content policy).
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 08:17 PM