Skip to main content

Security Audit

grpo-rl-training

github.com/davila7/claude-code-templates
AI SkillCommit 458b11867eae
15
CRITICAL
Scanned about 2 months ago
2
Critical
Immediate action required
1
High
Priority fixes suggested
1
Medium
Best practices review
1
Low
Acknowledged / Tracked

Trust Assessment

grpo-rl-training received a trust score of 15/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.

SkillShield's automated analysis identified 6 findings: 2 critical, 1 high, 1 medium, and 1 low severity. Key findings include Arbitrary command execution, Dangerous call: exec(), Network egress to untrusted endpoints.

The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Manifest Analysis layer scored lowest at 61/100, indicating areas for improvement.

Last analyzed on February 12, 2026 (commit 458b1186). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis
61%
Static Code Analysis
70%
Dependency Graph
100%
LLM Behavioral Safety
85%

Behavioral Risk Signals

Network Access
1 finding
Shell Execution
3 findings
Dynamic Code
3 findings

Security Findings6

SeverityFindingLayerLocation

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/3379cc17105f0345.svg)](https://skillshield.io/report/3379cc17105f0345)
SkillShield Badge

Powered by SkillShield