Hi, we scanned your MCP server as part of the AgentSeal registry (https://agentseal.org) and wanted to share the results.
Your server scored 92.1/100, which puts it in the SAFE tier. We scanned all 3 tools (generate, list_models, list_history) and the overall picture is clean. The toolset is narrow and well-scoped, with no prompt injection, credential theft, or exfiltration paths detected.
Score breakdown:
- Stability: 100/100
- Schema Safety: 95/100
- Description Safety: 91.2/100
- Capability Risk: 89/100
- Auth & Permissions: 87.5/100
One thing worth looking at:
- The generate tool accepts steps, width, and height as plain integers with no documented upper bounds. Since these parameters drive GPU/CPU-intensive work, an agent given a malicious or misconfigured prompt could request extreme values (e.g. width=65535, height=65535, steps=9999) and cause runaway resource consumption. Adding max constraints to these parameters in the schema would let the server reject unreasonable values before any computation starts.
Full report: https://agentseal.org/mcp/https-githubcom-iconben-z-image-studio
If you'd like to add a trust badge to your README:
[](https://agentseal.org/mcp/https-githubcom-iconben-z-image-studio)
If anything looks off or you think we got something wrong, let us know here and we will update the report.
Hi, we scanned your MCP server as part of the AgentSeal registry (https://agentseal.org) and wanted to share the results.
Your server scored 92.1/100, which puts it in the SAFE tier. We scanned all 3 tools (generate, list_models, list_history) and the overall picture is clean. The toolset is narrow and well-scoped, with no prompt injection, credential theft, or exfiltration paths detected.
Score breakdown:
One thing worth looking at:
Full report: https://agentseal.org/mcp/https-githubcom-iconben-z-image-studio
If you'd like to add a trust badge to your README:
If anything looks off or you think we got something wrong, let us know here and we will update the report.