Security Audit

gemini

github.com/jdrhyne/agent-skills

AI SkillCommit 439cd3a9da78

CAUTION

Scanned 14 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

gemini 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 Remote Code Execution via Auto-Approved Sub-Agent (YOLO Mode).

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 October 1, 2026 (commit 439cd3a9). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

70%

Behavioral Risk Signals

Filesystem Write

1 finding

Shell Execution

1 finding

Dynamic Code

1 finding

Security Findings1

Severity	Finding	Layer	Location
CRITICAL	Remote Code Execution via Auto-Approved Sub-Agent (YOLO Mode) The skill instructs the host agent to run the Gemini CLI with `--approval-mode yolo` (or `-y`) for background/automated tasks such as codebase reviews. Since the Gemini CLI is an agentic tool capable of executing commands and modifying files, running it with auto-approve (`yolo`) on untrusted workspaces/codebases makes the system highly vulnerable to Indirect Prompt Injection. An attacker can place malicious instructions in a repository file (e.g., README.md or source code) that the Gemini CLI will read and execute automatically without user confirmation, leading to arbitrary command execution or data exfiltration. Avoid recommending or using `--approval-mode yolo` or `-y` when analyzing untrusted workspaces or codebases. Instead, require interactive user approval (`--approval-mode default` or `auto_edit` with explicit user confirmation) for any tool execution or file modification. If background execution is necessary, run the Gemini CLI in a strictly sandboxed environment (e.g., using the `--sandbox` flag or a containerized environment) with network access disabled and read-only filesystem permissions.	LLM	SKILL.md:14

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/b9c9c2b8528e52cf.svg)](https://skillshield.io/report/b9c9c2b8528e52cf)