Security Audit
moosend-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
moosend-automation received a trust score of 74/100, 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 Access via RUBE_REMOTE_WORKBENCH, Potential Data Exfiltration via Generic Workbench Tool.
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
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Broad Tool Access via RUBE_REMOTE_WORKBENCH The skill explicitly recommends and provides an example for `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` for 'Bulk ops'. This tool is highly generic and can execute arbitrary Composio tools, potentially granting the LLM access to functionalities beyond Moosend automation, including other connected toolkits or system-level operations if available through Composio. This broad access exceeds the stated purpose of 'Moosend automation' and significantly increases the attack surface. Restrict the available tools to only Moosend-specific operations. If `RUBE_REMOTE_WORKBENCH` is necessary, ensure its scope is strictly limited to Moosend toolkit functions, or provide a more specific tool for bulk Moosend operations instead of a generic workbench. | LLM | SKILL.md:60 | |
| HIGH | Potential Data Exfiltration via Generic Workbench Tool The `RUBE_REMOTE_WORKBENCH` tool, recommended for 'Bulk ops', allows the execution of `run_composio_tool()`. This generic capability, combined with the potential for Moosend tools to read sensitive data (e.g., subscriber lists, campaign content, personal data), creates a credible path for data exfiltration. A malicious prompt could instruct the LLM to use `RUBE_REMOTE_WORKBENCH` to read sensitive Moosend data and then potentially exfiltrate it by passing it to another tool (if available via Composio) or by including it in the LLM's response. Implement strict access controls and data flow monitoring for `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()`. Ensure that tools capable of reading sensitive data cannot be chained with tools capable of external communication without explicit user consent or robust data sanitization. Consider providing a more specialized, data-exfiltration-resistant tool for bulk operations. | LLM | SKILL.md:60 |
Scan History
Embed Code
[](https://skillshield.io/report/51df92ddb8b743f9)
Powered by SkillShield