Trust Assessment
veo-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 1 finding: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Broad tool execution via Rube MCP allows excessive permissions.
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 Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Broad tool execution via Rube MCP allows excessive permissions The skill documentation explicitly encourages and demonstrates the use of `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH` to execute any tool discovered within the `veo` toolkit. This grants the LLM broad, unconstrained execution capabilities over all operations provided by the Veo toolkit through Rube MCP. An attacker who can manipulate the LLM's prompts could potentially execute arbitrary Veo operations, limited only by the scope of the Veo toolkit and the authenticated user's permissions. This design delegates significant power to the LLM without granular control within the skill itself. Implement granular access control for specific Veo operations rather than a generic 'execute any tool' mechanism. If broad execution is necessary, ensure robust input validation and strict user consent/authorization for sensitive operations. Consider defining a more restricted set of allowed `tool_slug` values or requiring explicit user confirmation for high-impact actions before execution. | LLM | SKILL.md:39 |
Scan History
Embed Code
[](https://skillshield.io/report/53651a68870846d3)
Powered by SkillShield