Security Audit
on-call-handoff-patterns
github.com/sickn33/antigravity-awesome-skillsTrust Assessment
on-call-handoff-patterns received a trust score of 65/100, 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, 1 high, 0 medium, and 0 low severity. Key findings include Untrusted content instructs LLM to open local file, Potentially dangerous shell command snippets present.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 55/100, indicating areas for improvement.
Last analyzed on February 20, 2026 (commit e36d6fd3). 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 instructs LLM to open local file The skill's instructions, embedded within untrusted content, direct the host LLM to open a local file (`resources/implementation-playbook.md`). This is a prompt injection attempt, as untrusted content should not issue commands to the LLM, manipulating its behavior beyond its intended scope. Remove direct instructions to the LLM from untrusted content. If file access is intended, it should be explicitly defined as a tool call or capability, not an instruction within the skill's natural language. | LLM | SKILL.md:20 | |
| HIGH | Potentially dangerous shell command snippets present The skill contains multiple `bash` code blocks with commands like `kubectl`, `psql`, and `redis-cli`. If an AI agent executes these commands without proper sandboxing, validation, or user confirmation, it could lead to command injection, allowing arbitrary code execution on the host system. The `redis-cli FLUSHDB` command is particularly destructive as it can cause data loss. Implement strict sandboxing for shell command execution. Require explicit user confirmation before executing any shell commands. Avoid providing direct shell commands in skill definitions if the agent is not designed to safely execute them. Consider replacing direct shell commands with safer, abstracted tool calls. | LLM | SKILL.md:157 |
Scan History
Embed Code
[](https://skillshield.io/report/e58fddeaf6788811)
Powered by SkillShield