MiniMax Vision
Analysis
The skill appears to do what it says—analyze user-sent images with MiniMax—but users should verify the local mcporter setup and understand images may be processed by the configured MCP service.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
env["PATH"] = "/Users/js/.local/bin:" + env.get("PATH", "") ... cmd = ["mcporter", "call", "minimax.understand_image", "--args", ...]The helper resolves and executes a local mcporter binary, preferring a hard-coded user-specific directory; this dependency is central to the stated purpose but its provenance should be checked.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
当用户发送图片或让你分析图片时,自动调用: mcporter call minimax.understand_image --args '{"prompt": "你要求的描述", "image_source": "图片路径"}'This shows that user image references are passed through an MCP call to MiniMax for image understanding; this is purpose-aligned, but images may contain private content.
