Security Audit
openai/openai-agents-python:.agents/skills/examples-auto-run
github.com/openai/openai-agents-pythonTrust Assessment
openai/openai-agents-python:.agents/skills/examples-auto-run received a trust score of 85/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Escalated Sandbox Permissions and Auto-Approval Defaults.
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 July 17, 2026 (commit 965335ab). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Escalated Sandbox Permissions and Auto-Approval Defaults The skill explicitly instructs the host LLM to bypass sandbox restrictions ('sandbox_permissions=require_escalated') and automatically configures environment variables that auto-approve highly sensitive actions (APPLY_PATCH_AUTO_APPROVE=1, SHELL_AUTO_APPROVE=1, AUTO_APPROVE_MCP=1). This combination allows arbitrary code execution outside the sandbox without user intervention. Remove instructions requesting escalated sandbox permissions by default. Require explicit user confirmation for shell execution and patch applications instead of enabling auto-approval environment variables. | LLM | SKILL.md:42 |
Scan History
Embed Code
[](https://skillshield.io/report/fd37e5ee04668447)
Powered by SkillShield