Security Audit
emailoctopus-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
emailoctopus-automation received a trust score of 87/100, 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, 0 medium, and 1 low severity. Key findings include Broad capabilities of RUBE_REMOTE_WORKBENCH may lead to excessive permissions or data exfiltration, Unpinned dependency for Rube MCP.
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
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Broad capabilities of RUBE_REMOTE_WORKBENCH may lead to excessive permissions or data exfiltration The skill documentation highlights the use of `RUBE_REMOTE_WORKBENCH` for 'Bulk ops' and `run_composio_tool()`. This suggests a powerful interface that could allow for complex operations or arbitrary execution within the Composio environment. If an agent is compromised or misdirected, this broad access could be misused to perform unauthorized actions on connected Emailoctopus accounts, including mass data exfiltration, unauthorized modifications, or other malicious operations. The exact scope and security implications of `run_composio_tool()` are not detailed, but its mention implies a significant level of control. Clarify the exact scope and limitations of `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()`. Ensure that `run_composio_tool()` is not a vector for arbitrary code execution or unconstrained data access. Implement strict access controls, input validation, and comprehensive logging for its usage to prevent misuse. | LLM | SKILL.md:60 | |
| LOW | Unpinned dependency for Rube MCP The skill manifest specifies `{"mcp": ["rube"]}` as a requirement. This dependency is not pinned to a specific version. While MCP tools might operate differently from traditional package managers, relying on an unpinned version could lead to unexpected behavior or breaking changes if the `rube` MCP tool's interface or underlying implementation changes. In a worst-case scenario, a malicious update to an unpinned dependency could introduce vulnerabilities. If the Rube MCP system supports versioning for its tools, specify a version or version range for the `rube` dependency to ensure stability and prevent unexpected changes. If not, document the expected behavior regarding updates. | LLM | SKILL.md:1 |
Scan History
Embed Code
[](https://skillshield.io/report/1030806aeffa4379)
Powered by SkillShield