Security Audit
winston-ai-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
winston-ai-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 Unpinned Rube MCP dependency, Broad tool execution capability via 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 | Unpinned Rube MCP dependency The skill's manifest specifies a dependency on the 'rube' MCP without a version constraint. This means the skill will always use the latest version of Rube MCP. A malicious or buggy update to the Rube MCP could introduce vulnerabilities or unwanted behavior into this skill without explicit review, posing a supply chain risk. Pin the Rube MCP dependency to a specific, known-good version (e.g., `{"mcp": ["rube@1.2.3"]}`) to prevent automatic updates that could introduce supply chain risks. Regularly review and update the pinned version. | LLM | SKILL.md | |
| MEDIUM | Broad tool execution capability via Rube MCP The skill leverages the Rube MCP, which exposes powerful generic execution tools like `RUBE_MULTI_EXECUTE_TOOL` and `RUBE_REMOTE_WORKBENCH`. Specifically, `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` suggests the ability to execute arbitrary Composio tools, not just those related to Winston AI. While the skill's stated purpose is 'Winston AI automation', its reliance on a generic MCP with broad execution primitives means it implicitly gains access to all toolkits available through Rube MCP, potentially exceeding the intended scope and granting excessive permissions to the LLM. If possible, restrict the Rube MCP's capabilities for this specific skill to only the `winston_ai` toolkit. Alternatively, ensure that the LLM's usage of `RUBE_REMOTE_WORKBENCH` is strictly confined to Winston AI operations through careful prompt engineering and monitoring. The Rube MCP itself should implement granular access controls to limit the scope of `run_composio_tool()`. | LLM | SKILL.md:68 |
Scan History
Embed Code
[](https://skillshield.io/report/5df20fa0fce691c8)
Powered by SkillShield