Security Audit

cloud-architect

github.com/sickn33/antigravity-awesome-skills

AI SkillCommit e36d6fd3b3f6

TRUSTED

Scanned about 6 hours ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

cloud-architect 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 Untrusted instruction to access local file.

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

85%

Behavioral Risk Signals

Filesystem Write

1 finding

Shell Execution

1 finding

Dynamic Code

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Untrusted instruction to access local file The skill definition contains an instruction within the untrusted input block that attempts to direct the host LLM to 'open' a local file (`resources/implementation-playbook.md`). This is a prompt injection attempt, as it tries to manipulate the LLM's behavior to perform an action (file access) that is not part of its default conversational flow. If the LLM has file system access, this could lead to unintended information disclosure or execution of further instructions from the accessed file. Remove or rephrase the instruction to avoid direct commands to the LLM to 'open' files. Instead, the skill should describe what information is available in `resources/implementation-playbook.md` and let the user decide to access it, or rely on the LLM's inherent knowledge base if such a file is meant to be part of its training data. If the intent is for the LLM to provide the content of that file, it should be explicitly defined as a tool call or a specific capability, not a direct instruction within the prompt.	LLM	SKILL.md:16

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/327c35878ced5645.svg)](https://skillshield.io/report/327c35878ced5645)