design ecommerce

WarnAudited by ClawScan on May 10, 2026.

Overview

The skill mostly matches its image-editing purpose, but its runner script handles user input in a way that could let a crafted image path or URL run local code.

Review this skill carefully before installing. Its image-processing behavior is mostly expected, but the unsafe JSON parsing in the shell runner should be fixed before processing untrusted image paths or URLs. If you use it anyway, only provide trusted inputs, use a dedicated API key, and verify the publisher/source.

Findings (5)

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.

What this means

A maliciously crafted image path or URL could potentially run commands locally with the agent's permissions.

Why it was flagged

INPUT_JSON comes from the --input-json argument and can contain user-supplied image paths or URLs. Embedding it directly inside a python3 -c program can let crafted input break out of the string and execute Python code.

Skill content
d = json.loads('''${INPUT_JSON}''')
Recommendation

Parse JSON through stdin or sys.argv instead of interpolating it into Python source, and add tests for quotes and other special characters in user-provided fields.

What this means

The agent may call the API and process/upload the selected image as soon as it has the required inputs.

Why it was flagged

The skill intentionally lets the agent invoke shell scripts with user-derived parameters once required inputs are collected. This is purpose-aligned, but it increases the importance of safe argument handling.

Skill content
Then execute immediately without asking for an extra confirmation turn.

bash __SKILL_DIR__/scripts/run_command.sh <action> --input-json '<params_json>'
Recommendation

Use the skill only for intended images and review prompts carefully; maintainers should ensure all shell arguments are safely escaped and validated.

What this means

The skill can use the configured Designkit account key and may consume credits for requested operations.

Why it was flagged

The skill requires a Designkit/OpenClaw API key and uses it across the included sub-skills. This is expected for the service, but it is still delegated account authority.

Skill content
"DESIGNKIT_OPENCLAW_AK": { "type": "string", "description": "OpenClaw/Designkit API Access Key. Shared by all sub-skills via header X-Openclaw-AK.", "required": true }
Recommendation

Use a dedicated API key if available, keep request logging off unless debugging, and do not share the key in chat or files.

What this means

Any selected local image may leave your machine for processing by the provider.

Why it was flagged

Local images may be sent to the external Designkit/OpenClaw service to complete the requested editing task. This is disclosed and purpose-aligned.

Skill content
The script automatically uploads local image files.
Recommendation

Only provide images you are comfortable uploading to the service, especially for private products, people, or unreleased designs.

What this means

It may be harder to verify that the package came from the claimed publisher or to audit changes over time.

Why it was flagged

The registry metadata does not identify a source repository for provenance, although the reviewed package includes its scripts and no install-time downloader is shown.

Skill content
Source: unknown
Recommendation

Verify the publisher and homepage before installing, and prefer a package with clear official source provenance.