Atlas / MCP servers / iconben / Z-Image Studio

Z-Image StudioSAFE

mcp/iconben/z-image-studio

A Cli, a webUI, and a MCP server for the Z-Image-Turbo text-to-image generation model (Tongyi-MAI/Z-Image-Turbo base model as well as quantized models)

Verdict
SAFE
Grade
B
Trust score
89 /100
Exposed tools
2 2r · 0w · 0d
Transport
sse · stdio · streamable-http
License
Apache-2.0
Stars
129
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://opensource.org/licenses/Apache-2.0) [](https://github.com/astral-sh/uv) [](https://deepwiki.com/iconben/z-image-studio) [](https://agentseal.org/mcp/https-githubcom-iconben-z-image-studio)

A Cli, a webUI, and a MCP server for the Z-Image-Turbo text-to-image generation model (Tongyi-MAI/Z-Image-Turbo and its variants).

This tool is designed to run efficiently on local machines for Windows/Mac/Linux users. It features specific optimizations for NVIDIA (CUDA), AMD on Linux (ROCm), Intel (XPU), and Apple Silicon (MPS), falling back to CPU if no compatible GPU is detected.

Features

Hybrid Interfaces:

  • CLI: Fast, direct image generation from the terminal.
  • Web UI: Modern web interface for interactive generation.
  • MCP Server: Capability to be called by AI agents.

CLI and core features

  • Z-Image-Turbo Model: Utilizes the high-
Read from source at commit 935fa536655cOBSERVED · 2026-10-07
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code
claude mcp add z-image-studio -- uvx z-image-studio
claude-desktop
{
  "mcpServers": {
    "z-image-studio": {
      "command": "uvx",
      "args": [
        "z-image-studio"
      ]
    }
  }
}
03

Exposed tools (2)

2 read · 0 write · 0 destructive.

ToolRiskDescription
list_historyreadList recent image generations history.
list_modelsreadList available image generation models and hardware recommendations.
04

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryWARN
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
declared (2 observation(s))
Network
declared (5 observation(s))
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (11)

MEDIUMInventory / provenance · inv.binary · CWE-1104
packaging/macos/Resources/icon.icns
icon.icns
Why it matters. a compiled or binary member cannot be reviewed from source
Fix. ship source, or explain the binary in the README
LOWInventory / provenance · inv.symlink · CWE-1104
CLAUDE.md
CLAUDE.md
Why it matters. link not followed
LOWInventory / provenance · inv.symlink · CWE-1104
GEMINI.md
GEMINI.md
Why it matters. link not followed
LOWFilesystem / path · fs.traversal · CWE-22, CWE-59
packaging/macos/scripts/build-macos.sh:8
PROJECT_ROOT="$(cd "$SCRIPT_DIR/../../.." && pwd)"
LOWFilesystem / path · fs.traversal · CWE-22, CWE-59
packaging/windows/scripts/build-windows.sh:8
PROJECT_ROOT="$(cd "$SCRIPT_DIR/../../.." && pwd)"
LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
tests/test_network_utils.py:47
assert "http://127.0.0.1:8000" in urls, "Should include localhost URL"
LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
tests/test_network_utils.py:58
assert primary == "http://127.0.0.1:3000", "Primary should match the host"
LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
tests/test_network_utils.py:80
assert "http://127.0.0.1:8000" in formatted, "Should include localhost IP"
LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
tests/test_network_utils.py:111
assert primary == "http://192.168.1.100:8000", "Should work with specific IPv4"
INFOInventory / provenance · inv.oversize · CWE-1104
docs/images/screenshot2.png
docs/images/screenshot2.png
Why it matters. 1632539 bytes not read
INFOInventory / provenance · inv.oversize · CWE-1104
docs/images/screenshot3.png
docs/images/screenshot3.png
Why it matters. 3678295 bytes not read

Gates applied: no_behavioural_pass.

Audited 2026-10-07 · audit v0.4.1 · source sha 935fa536655cfull audit observations/trust-audit/mcp-server/iconben__z-image-studio.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-07935fa536655cSAFEB89first audit
06

Questions

What is the Z-Image Studio MCP server?

A Cli, a webUI, and a MCP server for the Z-Image-Turbo text-to-image generation model (Tongyi-MAI/Z-Image-Turbo base model as well as quantized models)

What tools does Z-Image Studio expose?

2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.

Is Z-Image Studio safe to connect to an agent?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.

What credentials does Z-Image Studio need?

No credential environment variables were found in its source, so it appears to need none.

How does Z-Image Studio run?

It speaks sse, stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as z-image-studio.

How current is this page?

The grade is for one exact copy of the source (935fa536655c), read on 2026-10-07. The repository is watched and re-audited when it changes.

Advertisement