Security Audit

iqair-airvisual-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 27904475d127

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

iqair-airvisual-automation received a trust score of 82/100, 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 Generic Tool Execution via Rube MCP, Reliance on External Rube MCP for Tool Execution.

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

78%

Behavioral Risk Signals

Network Access

1 finding

Shell Execution

2 findings

Excessive Permissions

1 finding

Security Findings2

Severity	Finding	Layer	Location
HIGH	Generic Tool Execution via Rube MCP The skill exposes `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH` which are generic functions for executing arbitrary tools provided by the Rube MCP. While examples show usage for 'Iqair Airvisual operations', these tools themselves are not inherently scoped to a single toolkit. An LLM using this skill could be prompted to discover and execute tools from other connected toolkits via `RUBE_SEARCH_TOOLS` and then execute them using these generic execution functions, leading to broader access than intended. `RUBE_REMOTE_WORKBENCH` explicitly mentions `run_composio_tool()`, indicating its generic nature and potential for executing any Composio tool. Implement stricter scoping within the skill definition or the Rube MCP integration to ensure that `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH` can only execute tools explicitly related to the `iqair_airvisual` toolkit, or require explicit user confirmation for cross-toolkit execution.	LLM	SKILL.md:50
MEDIUM	Reliance on External Rube MCP for Tool Execution The skill's functionality is entirely dependent on the external Rube MCP (`https://rube.app/mcp`). The security and integrity of this skill are directly tied to the trustworthiness and security posture of the Rube MCP. A compromise of the Rube MCP could lead to malicious tool execution, data exfiltration, or other security incidents through this skill, as it acts as a central point for tool discovery and execution. While inherent to the design, mitigation strategies include: 1. Thoroughly vetting the Rube MCP provider and its security practices. 2. Implementing strong access controls and monitoring for activities performed via the Rube MCP. 3. Considering self-hosting or sandboxing the MCP if possible, or using a trusted, audited provider. 4. Ensuring the LLM's permissions are minimized even when interacting with the MCP.	LLM	SKILL.md:19

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/a3e0e180d6919b9a.svg)](https://skillshield.io/report/a3e0e180d6919b9a)