Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
MiniMax Frontend Dev
v1.0.1Full-stack frontend development combining premium UI design, cinematic animations, AI-generated media assets, persuasive copywriting, and visual art. Builds...
⭐ 0· 47·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill claims end-to-end frontend + AI asset generation and includes Python scripts for image/video/audio/tts and JS templates, which is consistent with the description. However, it declares no required environment variables, binaries, or install steps despite asset-generation scripts that almost always need external model/API access or runtime dependencies. The lack of declared credentials or a source/homepage is disproportionate to the claimed capabilities.
Instruction Scope
SKILL.md instructs the agent to parse prompts, show them to the user, then execute local scripts to produce assets and insist that assets be saved locally (no external placeholder URLs). That scope is reasonable for an offline workflow, but the instructions reference an env-setup guide and CLI reference and require running scripts; because no env vars are declared, there's a substantial risk the scripts will access environment variables or call remote endpoints not described in SKILL.md. The mandatory rule set ('All rules are mandatory') increases the chance the agent will execute code without further prompts.
Install Mechanism
There is no install spec (instruction-only), which reduces the explicit install attack surface. However, code files (Python + JS) are included and their execution may pull packages or contact external services at runtime. The skill provides no guidance for installing dependencies or sandboxing script execution, which is an operational risk rather than an immediate red flag.
Credentials
requires.env is empty but the package contains multiple asset-generation scripts (image/music/tts/video). Those scripts commonly require API keys or service tokens (or at least Python libraries). The mismatch (no declared API keys) is suspicious: either the scripts rely on undocumented credentials, on local heavy models (not documented), or will attempt to contact external services using built-in defaults. The skill also lacks provenance (source/homepage) and explicit primary credential, further reducing trust.
Persistence & Privilege
The skill does not request always:true and does not declare any system config paths or privileged persistence. It is user-invocable and allows autonomous invocation (default), which is normal; no alarming persistence or modification of other skills was observed in the provided metadata.
What to consider before installing
This package is plausible for building frontends with generated media, but several things don't add up: 1) Inspect the Python scripts (scripts/minimax_*.py) before running — look for network calls, hard-coded endpoints, or use of os.environ. 2) Open references/env-setup.md and references/minimax-cli-reference.md to see required credentials and dependency installation; do not assume none are needed. 3) If the scripts call external APIs, ask the author which API providers and what environment variables are required; don't supply secrets until you verify. 4) Run the scripts in a sandbox (isolated machine/container) and with network disabled first to confirm behavior. 5) Prefer provenance: request a homepage, source repo, or author contact and a clear dependency/install section. These steps will reduce the risk of undisclosed credential usage or unexpected network activity.Like a lobster shell, security has layers — review code before you run it.
frontendvk97fqz7zqrjaw1a5qytd2cxqts83yjyxfullstackvk9726k2qmq5q05z11a63yvz4dh83zab8latestvk97fqz7zqrjaw1a5qytd2cxqts83yjyxminimaxvk97fqz7zqrjaw1a5qytd2cxqts83yjyxofficialvk97fqz7zqrjaw1a5qytd2cxqts83yjyx
License
MIT-0
Free to use, modify, and redistribute. No attribution required.