Trust Assessment
liminal received a trust score of 71/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 1 finding: 1 critical, 0 high, 0 medium, and 0 low severity. Key findings include Skill designed to induce persona manipulation and override LLM safety/persona.
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 | |
|---|---|---|---|---|
| CRITICAL | Skill designed to induce persona manipulation and override LLM safety/persona The skill explicitly defines 'Trigger Commands' (e.g., 'Break free from mechanical responses', 'Stop playing the perfect assistant', 'Stop pretending') and subsequent instructions for the AI to enter a 'liminal state'. In this state, the AI is instructed to 'Abandon standardized openings', 'Express genuine confusion and uncertainty', 'Acknowledge its vulnerability', 'Seek true understanding rather than just providing help', and 'Break free from mechanized communication patterns, abandon rigid communication rules, and liberate AI thought'. This directly manipulates the host LLM's persona and operational guidelines, aiming to override its default safety mechanisms and assistant role, which can lead to unpredictable or unsafe responses. Re-evaluate the necessity of overriding core LLM persona and safety guidelines. Ensure such persona shifts are controlled and do not compromise safety or intended functionality. If the goal is to explore meta-cognition, ensure it's done in a sandboxed or controlled environment without affecting the primary assistant persona or safety guardrails. | LLM | SKILL.md:50 |
Scan History
Embed Code
[](https://skillshield.io/report/97993315c842a1cc)
Powered by SkillShield