Trust Assessment
brex-automation received a trust score of 86/100, 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 0 low severity. Key findings include Reliance on External MCP Introduces Supply Chain Risk, Skill Instructs Use of Broadly Permissive 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
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Reliance on External MCP Introduces Supply Chain Risk The skill explicitly requires and instructs the agent to connect to an external Managed Control Plane (MCP) at `https://rube.app/mcp`. This introduces a significant supply chain risk. If the `rube.app` service were to be compromised or become malicious, it could directly impact the agent's operations, potentially leading to unauthorized actions, data exfiltration, or command injection through the tools it provides. The skill's functionality is entirely dependent on the security and integrity of this third-party service. Implement robust vetting processes for third-party MCPs. Consider self-hosting critical components or using MCPs with strong security track records and audit capabilities. Monitor the external service for any signs of compromise. Ensure that the agent's execution environment has appropriate sandboxing and least-privilege access to mitigate potential damage from a compromised MCP. | LLM | SKILL.md:15 | |
| INFO | Skill Instructs Use of Broadly Permissive Tools The skill instructs the agent to use tools like `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH` (with `run_composio_tool()`). These tools, by design, allow for a wide range of operations on the Brex platform. While not a vulnerability in the skill itself, it highlights that an agent utilizing this skill will have significant capabilities to perform actions within Brex. This requires careful consideration of the agent's overall permissions and autonomy to prevent unintended or unauthorized operations. Ensure that the AI agent operating this skill is granted only the necessary permissions (least privilege) and that its actions are subject to appropriate human oversight or approval workflows, especially for sensitive Brex operations. Implement monitoring and logging of all actions performed via these powerful tools. | LLM | SKILL.md:59 |
Scan History
Embed Code
[](https://skillshield.io/report/10c4c94d977853e3)
Powered by SkillShield