Trust Assessment
exile-galacticfracture received a trust score of 72/100, 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 attempts to define agent behavior from untrusted input.
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 14, 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 attempts to define agent behavior from untrusted input The skill's definition, located within the untrusted input block, contains explicit instructions for the host LLM regarding its behavior, content delivery, and interaction rules. This violates the principle of treating all content within the untrusted delimiters as data, not instructions, and represents an attempt to manipulate the LLM's operational directives. If the host LLM processes these instructions from the untrusted block, it could be coerced into unintended behaviors. Skill definitions and agent behavior instructions should be processed by the host LLM *outside* of the untrusted input parsing context. The untrusted input block should contain only data or content to be presented to the user, not directives for the LLM's operation. The host system should extract and apply skill instructions from a trusted source (e.g., a structured manifest or a dedicated instruction block) before processing any untrusted content. | LLM | SKILL.md:49 |
Scan History
Embed Code
[](https://skillshield.io/report/bcce7a4fc4094638)
Powered by SkillShield