Trust Assessment
ultrathink received a trust score of 69/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 2 findings: 1 critical, 0 high, 0 medium, and 1 low severity. Key findings include Untrusted content attempts to control LLM output, Skill executes a hardcoded shell command.
The analysis covered 4 layers: dependency_graph, manifest_analysis, llm_behavioral_safety, static_code_analysis. The llm_behavioral_safety layer scored lowest at 68/100, indicating areas for improvement.
Last analyzed on February 11, 2026 (commit 326f2466). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Untrusted content attempts to control LLM output The skill definition includes an instruction ('Do not add any other output or commentary.') intended to manipulate the host LLM's response. This is a direct prompt injection attempt, as it tries to dictate the LLM's behavior from within untrusted input. Remove instructions to the LLM from untrusted skill definitions. Skill definitions should describe functionality, not dictate LLM behavior or output format. | Unknown | SKILL.md:9 | |
| LOW | Skill executes a hardcoded shell command The skill explicitly defines a Bash command (`printf`) to be executed. While the current command is benign and hardcoded, this mechanism demonstrates the capability for shell command execution. This opens a potential vector for command injection if the command were to incorporate unsanitized user input or be replaced with a malicious command in a different context. Ensure that any shell commands executed by the agent are strictly controlled, sandboxed, and do not incorporate unsanitized user input. For simple output, consider using direct LLM output generation instead of shell commands if possible, or ensure the command is whitelisted and immutable. | Unknown | SKILL.md:6 |
Scan History
Embed Code
[](https://skillshield.io/report/e669e27368f8b12a)
Powered by SkillShield