Security Audit

garrytan/gstack:openclaw/skills/gstack-openclaw-office-hours

github.com/garrytan/gstack

AI SkillCommit e8893a18b18e

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

garrytan/gstack:openclaw/skills/gstack-openclaw-office-hours received a trust score of 78/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.

SkillShield's automated analysis identified 2 findings: 0 critical, 1 high, 1 medium, and 0 low severity. Key findings include Broad Data Access and Potential Exfiltration, Excessive Filesystem Permissions Requested.

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 April 29, 2026 (commit e8893a18). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

78%

Dependency Graph

100%

LLM Behavioral Safety

100%

Behavioral Risk Signals

Network Access

1 finding

Filesystem Write

2 findings

Excessive Permissions

2 findings

Security Findings2

Severity	Finding	Layer	Location
HIGH	Broad Data Access and Potential Exfiltration The skill explicitly instructs the agent to 'Read the workspace and any existing project docs', 'Check git log', and 'Search the codebase'. This grants very broad read access to potentially sensitive user data, including source code, configuration files, and historical changes. If the agent is compromised or if the skill's output is not carefully controlled, this information could be exfiltrated. Additionally, the instruction to 'Save the design doc to memory/' implies write access to a persistent storage location, which could be used to store exfiltrated data or malicious payloads. Restrict the agent's access to the filesystem to only necessary directories and file types. Implement strict output filtering and sanitization to prevent sensitive data from being included in the agent's responses. For 'memory/' storage, ensure it's sandboxed and its contents are regularly reviewed or ephemeral.	Static	SKILL.md:20
MEDIUM	Excessive Filesystem Permissions Requested The skill's instructions to 'Read the workspace', 'Check git log', and 'Search the codebase' imply a requirement for extensive filesystem read permissions across the entire project directory. While intended for context gathering, this level of access is excessive for a skill primarily focused on design and brainstorming, increasing the attack surface. The instruction to save to 'memory/' also requires write permissions. Refine the scope of required filesystem access. Instead of 'Read the workspace', specify particular directories or file types. If git log is needed, consider providing a more constrained interface or summary rather than full access. Ensure write access to 'memory/' is limited to the skill's specific needs and within a secure, isolated directory.	Static	SKILL.md:20

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/a42989805cf3e1d8.svg)](https://skillshield.io/report/a42989805cf3e1d8)