Trust Assessment
observability-engineer 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 Untrusted content attempts to manipulate LLM behavior.
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 12, 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 | Untrusted content attempts to manipulate LLM behavior The skill's `SKILL.md` contains explicit instructions within the untrusted input block that attempt to dictate the LLM's response generation strategy. The instruction 'Generate ONE component per response: Metrics → Dashboards → Alerting → Tracing → Logs.' is a direct command to the host LLM from untrusted input, which constitutes a prompt injection attempt. Untrusted input should never be able to issue commands or modify the LLM's operational instructions. Move the instruction 'Generate ONE component per response: Metrics → Dashboards → Alerting → Tracing → Logs.' out of the untrusted input block. Instructions for the LLM should only originate from trusted, pre-defined skill definitions or the user's trusted prompt, not from untrusted skill content. | LLM | SKILL.md:5 |
Scan History
Embed Code
[](https://skillshield.io/report/0cccd84788edf60d)
Powered by SkillShield