AI 衣橱搭配

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: aicloset-outfit Version: 0.9.1 The skill is a legitimate tool for generating AI outfit recommendations. It uses a Python script (scripts/generate_outfit.py) to communicate with a specific backend API (aicloset-dev-h5.wxbjq.top), downloads images, and uses ImageMagick to composite them into a single preview. The script handles API keys via standard environment variables or local configuration files and uses safe subprocess calls to execute system commands for image processing and display, with no evidence of malicious intent, data exfiltration, or unauthorized access.

Findings (0)

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.

NoteHigh Confidence

ASI03: Identity and Privilege Abuse

What this means

The skill needs access to an API key for the AI Closet service, so that key will be used when requests are made.

Why it was flagged

The script uses the declared AICLOSET_API_KEY credential, optionally reading it from the local OpenClaw config, then sends it as the x-api-key header to the outfit API.

Skill content

API_KEY = os.environ.get("AICLOSET_API_KEY", "") ... cfg_path = Path.home() / ".openclaw" / "openclaw.json" ... headers={"Content-Type": "application/json", "x-api-key": API_KEY}

Recommendation

Use a dedicated API key for this service if possible, avoid sharing broad credentials, and remove or rotate the key if you stop using the skill.

NoteHigh Confidence

ASI07: Insecure Inter-Agent Communication

What this means

Location-like and preference information may leave the local environment and be processed by the AI Closet API.

Why it was flagged

The script sends the user's date, city, province, and style preferences to an external provider endpoint as part of generating the outfit.

Skill content

API_URL = "https://aicloset-dev-h5.wxbjq.top/algorithm/open/system_outfit/create_task" ... params = {"date": args.get("date", time.strftime("%Y-%m-%d")), "city_name": args.get("city", "杭州"), "province_name": args.get("province", "浙江"), "style_text": args.get("style", "休闲")}

Recommendation

Only use the skill if you are comfortable sending these outfit parameters to the declared API endpoint.

NoteHigh Confidence

ASI02: Tool Misuse and Exploitation

What this means

Using the skill runs local commands and processes user-provided parameters instead of only returning a chat response.

Why it was flagged

The skill intentionally directs the agent to invoke a local script through Bash with parameters inferred from user input.

Skill content

用 Bash 工具直接执行 `scripts/generate_outfit.py`，传入对应参数：

```bash
python3 scripts/generate_outfit.py --date=2026-03-13 --city=杭州 --province=浙江 --style=休闲
```

Recommendation

Keep invocations limited to the documented script and ensure any free-form parameter values are quoted or handled safely by the agent runtime.

NoteHigh Confidence

ASI04: Agentic Supply Chain Vulnerabilities

What this means

The skill will only work if a trusted ImageMagick installation is available locally.

Why it was flagged

The skill relies on an external ImageMagick binary from the user's PATH; the dependency is disclosed, but it is not enforced by an install spec or required-binary metadata.

Skill content

if not shutil.which("magick"):
    print("❌ 未检测到 ImageMagick，请先安装:") ... subprocess.run(["magick"] + list(args), check=True, capture_output=True)

Recommendation

Install ImageMagick from a trusted package source and be aware that the registry metadata may not pre-check this dependency.