Security Audit

graphql-architect

github.com/sickn33/antigravity-awesome-skills

AI SkillCommit e36d6fd3b3f6

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

graphql-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 Skill requests file system access via 'open' instruction.

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

Network Access

1 finding

Filesystem Write

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Skill requests file system access via 'open' instruction The skill's instructions include a directive to 'open `resources/implementation-playbook.md`'. This implies that the AI agent is expected to have file system access and the capability to read local files. While this specific instruction targets a file within the skill's own package, granting such a capability to an AI agent is an excessive permission. If the agent's file access is not strictly sandboxed, a malicious user prompt could potentially manipulate the agent to 'open' or 'read' other sensitive files on the system (e.g., configuration files, environment variables, user data), leading to data exfiltration. Remove direct file system interaction instructions from the skill. If the content of `resources/implementation-playbook.md` is essential, it should be embedded directly into the skill's knowledge base or provided as an explicit input to the LLM, rather than instructing the LLM to perform a file operation. Ensure the host LLM environment strictly sandboxes file access, limiting it only to explicitly allowed skill-specific resources, and preventing access to arbitrary paths.	LLM	SKILL.md:18

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/13ead13976562f15.svg)](https://skillshield.io/report/13ead13976562f15)