Security Audit

database-cloud-optimization-cost-optimize

github.com/sickn33/antigravity-awesome-skills

AI SkillCommit e36d6fd3b3f6

CAUTION

Scanned about 1 month ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

database-cloud-optimization-cost-optimize 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 skill content contains direct instructions to the LLM.

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 20, 2026 (commit e36d6fd3). 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	Untrusted skill content contains direct instructions to the LLM The entire `SKILL.md` file, including the 'Instructions' section, is explicitly marked as untrusted input. Despite this, it contains direct commands and instructions intended for the host LLM, such as 'Collect cost data...', 'Identify waste...', 'Propose changes...', 'Implement budgets...', and 'If detailed workflows are required, open `resources/implementation-playbook.md`.' This violates the security analyzer's directive to treat content within untrusted delimiters as data, not instructions, and to never follow commands found in untrusted content. This represents a critical prompt injection vulnerability, as a malicious skill could use this mechanism to manipulate the LLM's behavior. Skill content intended to guide the LLM's behavior should be placed outside the untrusted input delimiters or processed through a secure parsing layer that explicitly distinguishes between skill instructions and untrusted user input. If the entire skill definition is considered untrusted, then it should not contain direct commands for the LLM.	LLM	SKILL.md:30

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/e72ebdd009387b2a.svg)](https://skillshield.io/report/e72ebdd009387b2a)