Security Audit

memberstack-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 27904475d127

CAUTION

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

memberstack-automation received a trust score of 73/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.

SkillShield's automated analysis identified 2 findings: 0 critical, 2 high, 0 medium, and 0 low severity. Key findings include Broad Tool Execution Capability via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH, Agent Instructed to Follow External Authentication Links.

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 27904475). 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

Network Access

1 finding

Shell Execution

1 finding

Dynamic Code

1 finding

Excessive Permissions

2 findings

Security Findings2

Severity	Finding	Layer	Location
HIGH	Broad Tool Execution Capability via RUBE_MULTI_EXECUTE_TOOL and RUBE_REMOTE_WORKBENCH The skill instructs the AI agent to use `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH` which allow execution of arbitrary tools discovered via `RUBE_SEARCH_TOOLS`. This grants the agent broad permissions to perform any operation available through the Memberstack toolkit via Rube MCP. If an attacker can influence the `tool_slug` or `arguments` provided to these functions, they could execute unauthorized actions on the Memberstack account or other connected services. The `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` further suggests a general remote execution capability, increasing the risk of command injection against the remote service. Implement strict input validation and allowlisting for `tool_slug` and `arguments` based on user intent and predefined safe operations. Ensure the agent's execution environment for `RUBE_REMOTE_WORKBENCH` is sandboxed and has minimal necessary permissions. Consider requiring explicit user confirmation for sensitive operations before executing tools.	LLM	SKILL.md:59
HIGH	Agent Instructed to Follow External Authentication Links The skill explicitly instructs the AI agent to 'follow the returned auth link to complete setup' when managing connections via `RUBE_MANAGE_CONNECTIONS`. An automated agent following arbitrary external links, especially for authentication, poses a significant risk. A malicious or compromised link could lead to credential harvesting, session hijacking, or unauthorized access to the Memberstack account. The agent might not be able to discern legitimate authentication flows from phishing attempts, leading to credential exposure or data exfiltration. Agents should never automatically follow external authentication links. Instead, the agent should return the authentication link to the human user for manual completion. The agent should then be provided with the resulting credentials or confirmation token securely, without direct interaction with the external authentication flow.	LLM	SKILL.md:29

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/cca8e12893047968.svg)](https://skillshield.io/report/cca8e12893047968)