Trust Assessment
task 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 Potential Command Injection via tasker_cmd 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 | Potential Command Injection via tasker_cmd arguments The skill instructs the LLM to construct command-line arguments for the `tasker_cmd` tool by directly embedding untrusted user input (e.g., `<task>`, `<details>`, `<query>`). For example, the instruction `add "<task>" --today` takes user-provided text and places it directly into a command argument. If `tasker_cmd` is implemented in a way that executes these arguments via a shell (e.g., `subprocess.run(..., shell=True)`) or otherwise processes them without proper sanitization/escaping, a malicious user could inject arbitrary shell commands. The `command-arg-mode: raw` setting in the manifest further increases this risk by allowing the LLM to generate the full argument string. 1. **Sanitize/Escape User Input:** Ensure that any user-provided strings embedded into `tasker_cmd` arguments are properly sanitized or shell-escaped before being passed to the tool. 2. **Tool Implementation Review:** Review the `tasker_cmd` tool's implementation to ensure it does not use `shell=True` for `subprocess.run` or similar functions, and that it robustly handles untrusted input, treating arguments as data, not executable code. 3. **Argument Validation:** Implement strict validation of user input before constructing commands. | LLM | SKILL.md:25 |
Scan History
Embed Code
[](https://skillshield.io/report/66b804cb20bc7578)
Powered by SkillShield