Security Audit

zerobounce-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 99e2a2951545

TRUSTED

Scanned 6 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

zerobounce-automation received a trust score of 95/100, placing it in the Trusted category. This skill has passed all critical security checks and demonstrates strong security practices.

SkillShield's automated analysis identified 1 finding: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Broad Rube MCP access via generic execution tools.

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 17, 2026 (commit 99e2a295). 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

93%

Behavioral Risk Signals

Shell Execution

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
MEDIUM	Broad Rube MCP access via generic execution tools The skill requires access to Rube MCP (`mcp: ['rube']`). While the skill's stated purpose is Zerobounce automation, it instructs the LLM to use generic Rube MCP execution tools such as `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH`. These tools can interact with any toolkit configured under Rube MCP, not just Zerobounce. This means the skill implicitly grants the LLM access to all other services connected via Rube MCP, which is a broader permission scope than the skill's specific Zerobounce functionality suggests. A malicious prompt could potentially leverage this to interact with other sensitive systems if they are also connected to Rube MCP through Rube MCP's generic execution capabilities. If possible, the Rube MCP integration should be scoped to specific toolkits (e.g., `mcp: ['rube:zerobounce']`) to limit the LLM's access. Alternatively, the skill's instructions should explicitly guide the LLM to only use Zerobounce-specific tool slugs when calling generic execution tools like `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH`. The Rube MCP system itself could also implement more granular permissions at the toolkit level for LLM integrations.	LLM	SKILL.md:70

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/2281705bb7497b4f.svg)](https://skillshield.io/report/2281705bb7497b4f)