Trust Assessment
setup-automatik received a trust score of 10/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 5 findings: 4 critical, 1 high, 0 medium, and 0 low severity. Key findings include Arbitrary command execution, Remote code execution: curl/wget pipe to shell, Skill requires direct command execution on user's VPS.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 25/100, indicating areas for improvement.
Last analyzed on February 12, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings5
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Remote code download piped to interpreter Review all shell execution calls. Ensure commands are static (not built from user input), use absolute paths, and are strictly necessary. Prefer library APIs over shell commands. | Manifest | skills/alltomatos/setup-automatik/SKILL.md:38 | |
| CRITICAL | Remote code execution: curl/wget pipe to shell Detected a pattern that downloads and immediately executes remote code. This is a primary malware delivery vector. Never pipe curl/wget output directly to a shell interpreter. | Static | skills/alltomatos/setup-automatik/SKILL.md:38 | |
| CRITICAL | Skill requires direct command execution on user's VPS The skill's design explicitly requires the agent to have direct command execution capabilities on the user's Virtual Private Server (VPS) via 'OpenClaw Node Pairing'. This involves the user running a `curl | sh` command to install an agent that allows the LLM to execute arbitrary commands. This grants the agent extremely broad, potentially root-level, permissions on the user's server, creating a critical command injection vulnerability if the agent is compromised or manipulated. Re-evaluate the skill's architecture to minimize the need for direct, unconstrained command execution on user systems. Implement a more granular permission model or use sandboxed execution environments. If direct execution is unavoidable, ensure robust input validation and strict command whitelisting. Educate users on the risks of granting such broad access. | LLM | SKILL.md:42 | |
| CRITICAL | Skill requests sensitive SSH credentials from user The skill explicitly instructs the user to provide highly sensitive SSH connection details, including 'Public IP Address', 'Username (usually `root`)', and 'SSH Password OR Private Key'. Providing these credentials to an AI agent creates a severe risk of credential harvesting and data exfiltration. A compromised agent or a malicious prompt injection could lead to unauthorized access to the user's VPS. Never request raw SSH credentials (passwords or private keys) from users. Instead, implement secure authentication mechanisms like temporary, scoped API tokens, SSH keys managed by a secure vault, or OAuth flows. If SSH access is necessary, guide users to configure specific, restricted SSH keys for the agent with minimal permissions and IP restrictions. | LLM | SKILL.md:47 | |
| HIGH | Skill recommends `curl | sh` for installation, posing supply chain risk The skill recommends installing the OpenClaw agent using `curl -fsSL https://get.openclaw.ai | sh`. This method is inherently risky as it pipes the output of a remote script directly into a shell for execution. If the `get.openclaw.ai` domain or the script hosted there is compromised, users could unknowingly execute malicious code on their VPS, leading to a supply chain attack. Avoid recommending `curl | sh` for software installation. Instead, provide installation instructions that involve package managers (e.g., `apt`, `yum`), signed binaries, or verified container images. If a script must be used, advise users to download, review, and then execute it, or provide a checksum for verification. | LLM | SKILL.md:42 |
Scan History
Embed Code
[](https://skillshield.io/report/0e13399a6f8c43f0)
Powered by SkillShield