Trust Assessment
openclaw-security 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 9 findings: 5 critical, 4 high, 0 medium, and 0 low severity. Key findings include Arbitrary command execution, Dangerous call: subprocess.run(), Arbitrary Code Execution via User-Controlled Workspace Path.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Manifest Analysis layer scored lowest at 10/100, indicating areas for improvement.
Last analyzed on February 14, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings9
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Python shell execution (os.system, subprocess) 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/atlaspa/openclaw-security/scripts/security.py:207 | |
| CRITICAL | Arbitrary command execution Python shell execution (os.system, subprocess) 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/atlaspa/openclaw-security/scripts/security.py:284 | |
| CRITICAL | Arbitrary command execution Python shell execution (os.system, subprocess) 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/atlaspa/openclaw-security/scripts/security.py:612 | |
| CRITICAL | Arbitrary Code Execution via User-Controlled Workspace Path The `run_skill` function executes Python scripts from paths derived from the `--workspace` argument. If an attacker can control the `--workspace` argument (e.g., by setting the `OPENCLAW_WORKSPACE` environment variable or providing it directly), they can point the skill to a malicious directory. The orchestrator will then execute any Python script placed at the expected `skills/<skill_name>/<script_rel>` path within that malicious workspace. This allows for arbitrary code execution. Restrict the `--workspace` argument to trusted paths or implement strict validation to prevent arbitrary directory traversal. Consider signing skills and verifying signatures before execution, or running skills in a sandboxed environment. | LLM | scripts/security.py:200 | |
| CRITICAL | Unverified Skill Downloads from Unpinned Remote Source The `install_skill` function downloads skill packages directly from `https://raw.githubusercontent.com/openclaw/clawhub/main/skills` without any version pinning or integrity verification (e.g., checksums, cryptographic signatures). This makes the system vulnerable to supply chain attacks. If the `openclaw/clawhub` repository is compromised, or if the `main` branch is tampered with, malicious code could be downloaded and subsequently executed by the orchestrator. Implement cryptographic signature verification for all downloaded skill packages. Pin skill versions to specific commits or releases. Consider using a trusted, private skill registry with robust security controls. | LLM | scripts/security.py:230 | |
| HIGH | Dangerous call: subprocess.run() Call to 'subprocess.run()' detected in function 'run_skill'. This can execute arbitrary code. Avoid using dangerous functions like exec/eval/os.system. Use safer alternatives. | Static | skills/atlaspa/openclaw-security/scripts/security.py:207 | |
| HIGH | Dangerous call: subprocess.run() Call to 'subprocess.run()' detected in function 'cmd_install'. This can execute arbitrary code. Avoid using dangerous functions like exec/eval/os.system. Use safer alternatives. | Static | skills/atlaspa/openclaw-security/scripts/security.py:284 | |
| HIGH | Dangerous call: subprocess.run() Call to 'subprocess.run()' detected in function 'cmd_update'. This can execute arbitrary code. Avoid using dangerous functions like exec/eval/os.system. Use safer alternatives. | Static | skills/atlaspa/openclaw-security/scripts/security.py:612 | |
| HIGH | Broad Filesystem Access via User-Controlled Workspace Path The skill allows the `--workspace` argument to be set to an arbitrary path, which then dictates where skills are installed and from where they are executed. This grants the skill broad filesystem access, potentially allowing it to read, write, or execute files in sensitive system locations if the user provides a malicious path and the underlying operating system permissions allow it. This is exacerbated by the command injection vulnerability (SS-CMD-001). Implement strict validation for the `--workspace` argument, ensuring it points only to designated, sandboxed, or user-owned directories. Avoid resolving paths that could lead outside a safe root. | LLM | scripts/security.py:120 |
Scan History
Embed Code
[](https://skillshield.io/report/d2e34a54b1a4137a)
Powered by SkillShield