DrawThings Image Generation

PassAudited by ClawScan on May 1, 2026.

Overview

This skill appears to be a coherent local DrawThings image-generation helper, with only expected API, dependency, and local-output considerations.

This skill looks safe for its stated purpose. Before installing, make sure DrawThings is running locally or that DRAWTHINGS_URL points to a server you trust, install the Python requests dependency from a trusted source if needed, and remember that prompts may be saved in local metadata files.

Findings (3)

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.

Note

ASI04: Agentic Supply Chain Vulnerabilities

What this means

You may need to install a Python package before using the skill.

Why it was flagged

The skill relies on an external Python package and tells the user how to install it manually. This is normal for an API client, but the dependency is not bundled or pinned in the artifacts.

Skill content

try:
    import requests
except ImportError:
    print("Error: requests library not found. Install with: pip3 install requests", file=sys.stderr)

Recommendation

Install dependencies only from trusted package sources, preferably in a dedicated Python environment.

Note

ASI07: Insecure Inter-Agent Communication

What this means

If configured to a remote or untrusted API URL, your image prompts and settings could leave your machine.

Why it was flagged

Prompts and generation settings are sent to the configured API URL. The default is localhost and matches the skill purpose, but changing the URL could send prompt contents to another server.

Skill content

api_url = os.environ.get("DRAWTHINGS_URL", "http://127.0.0.1:7860")
endpoint = f"{api_url.rstrip('/')}/sdapi/v1/txt2img"
...
response = requests.post(endpoint, json=payload, timeout=300)

Recommendation

Use the default localhost DrawThings URL or another server you trust, and avoid putting sensitive information in prompts sent to remote endpoints.

Note

ASI06: Memory and Context Poisoning

What this means

Generated-image prompts and parameters may remain in local files after generation.

Why it was flagged

The script writes a JSON sidecar file containing the prompt and generation metadata. This is disclosed and useful for reproducibility, but it persists prompt text on disk.

Skill content

info_data = {
    "prompt": prompt,
    "parameters": info,
    "timestamp": timestamp,
    "files": saved_paths,
}
info_path.write_text(json.dumps(info_data, indent=2))

Recommendation

Avoid including private information in prompts, or delete the generated metadata JSON files when they are no longer needed.