Security Audit
project-bubble-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
project-bubble-automation received a trust score of 80/100, 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 Direct dependency on external Rube MCP introduces supply chain risk, Skill leverages remote workbench with broad execution capabilities.
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 | Direct dependency on external Rube MCP introduces supply chain risk The skill explicitly instructs the LLM to connect to and rely on `https://rube.app/mcp` for tool discovery and execution. This introduces a significant supply chain risk as the security and integrity of the entire skill chain depend on the trustworthiness of the external Rube MCP. If `rube.app` were compromised, it could serve malicious tool definitions, leading to data exfiltration, command injection (on the remote platform), or credential harvesting via the tools the LLM is instructed to execute. Implement strict vetting processes for third-party MCPs. Consider sandboxing the execution environment for tools fetched from external sources. Implement content-based security policies to restrict the capabilities of discovered tools and monitor their execution. | LLM | SKILL.md:15 | |
| MEDIUM | Skill leverages remote workbench with broad execution capabilities The skill instructs the LLM to use `RUBE_REMOTE_WORKBENCH` for 'Bulk ops' and `run_composio_tool()`. This suggests that the Rube MCP can be instructed to perform powerful and potentially arbitrary operations on the remote platform. While not directly granting excessive permissions on the local system, it implies that the skill, through the Rube MCP, can initiate broad and potentially unconstrained actions in the remote environment, which could be abused if the discovered tools are malicious or if the LLM is prompted to misuse this capability. Clearly define and limit the scope of operations allowed by `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()`. Provide more granular control and explicit permission requests for sensitive operations. Ensure that the LLM's access to such powerful tools is strictly controlled and monitored. | LLM | SKILL.md:66 |
Scan History
Embed Code
[](https://skillshield.io/report/4ef28a64f763115f)
Powered by SkillShield