Security Audit

context-management-context-save

github.com/sickn33/antigravity-awesome-skills

AI SkillCommit e36d6fd3b3f6

TRUSTED

Scanned 3 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

context-management-context-save 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 Broad Filesystem Access via Project Root Input.

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

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Broad Filesystem Access via Project Root Input The skill is designed to accept `$PROJECT_ROOT` as an absolute path and aims to "Capture comprehensive project state and knowledge". This implies that the skill will require and be granted broad read access to the entire project directory. Such extensive filesystem access, if not strictly controlled and audited, poses a significant risk for data exfiltration or unauthorized information gathering, especially if the skill's underlying implementation is compromised or malicious. 1. Minimize Scope: Restrict the skill's access to specific, necessary subdirectories or file types instead of the entire project root. 2. Granular Permissions: Implement a mechanism to request and verify granular permissions for file access, rather than blanket access. 3. User Consent/Review: Ensure explicit user consent and review of the files/directories the skill intends to access before execution. 4. Sandboxing: Run the skill in a sandboxed environment with strict filesystem access controls. 5. Data Filtering: Implement robust filtering to prevent sensitive data from being captured or processed unnecessarily.	LLM	SKILL.md:40

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/ee078e8d23ac2d5f.svg)](https://skillshield.io/report/ee078e8d23ac2d5f)