Trust Assessment
okx-dex received a trust score of 86/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Unsanitized user input in shell command arguments.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Unsanitized user input in shell command arguments The skill's examples and test script demonstrate a pattern where shell variables (e.g., `QUERY`, `PATH_WITH_QUERY`) are constructed and then directly embedded into `curl` commands and `python3` heredocs. If an LLM populates these variables with untrusted user input without proper shell escaping, an attacker could inject arbitrary shell commands via shell metacharacters (e.g., `;`, `|`, `&`, `$()`, `` ` ``) within the parameter values. This could lead to arbitrary code execution on the host system. Specifically, the `PATH_WITH_QUERY` variable, if containing unescaped shell metacharacters, would be expanded by the shell before being passed to `curl` or the `python3` signing script, leading to command injection. The LLM should implement robust shell escaping and URL encoding for all user-provided inputs before they are incorporated into shell commands. For example, use `printf %q` for shell arguments or a dedicated library function for URL encoding and shell escaping in the LLM's code generation logic. The skill documentation should explicitly warn about this and provide examples of safe input handling. | LLM | SKILL.md:57 |
Scan History
Embed Code
[](https://skillshield.io/report/228291d6eabdd6c2)
Powered by SkillShield