Security Audit
chatwork-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
chatwork-automation received a trust score of 68/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 2 findings: 1 critical, 0 high, 1 medium, and 0 low severity. Key findings include `RUBE_REMOTE_WORKBENCH` tool suggests arbitrary code execution capabilities, `RUBE_MULTI_EXECUTE_TOOL` allows execution of any discovered Chatwork operation.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 63/100, indicating areas for improvement.
Last analyzed on February 20, 2026 (commit 27904475). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | `RUBE_REMOTE_WORKBENCH` tool suggests arbitrary code execution capabilities The skill documentation mentions `RUBE_REMOTE_WORKBENCH` for 'Bulk ops' and refers to `run_composio_tool()`. The term 'workbench' and the nature of 'bulk operations' often imply broad access to the underlying environment, potentially including arbitrary code execution or shell commands. Without explicit sandboxing or scope limitations described, this tool presents a significant risk for command injection, data exfiltration, or unauthorized system access if a malicious prompt can instruct the LLM to use it. 1. Clarify the exact capabilities and limitations of `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()`. 2. If it allows arbitrary code execution, implement strict sandboxing and permission controls. 3. If not intended for arbitrary code, rename or re-document to avoid implying such capabilities. 4. Ensure that the LLM cannot be prompted to use this tool for unintended purposes. | LLM | SKILL.md:80 | |
| MEDIUM | `RUBE_MULTI_EXECUTE_TOOL` allows execution of any discovered Chatwork operation The `RUBE_MULTI_EXECUTE_TOOL` is designed to execute any tool slug discovered via `RUBE_SEARCH_TOOLS`. While this is the intended functionality of an automation skill, it means the skill, if prompted maliciously, could perform any operation available in the Chatwork toolkit without further explicit user confirmation for each specific action. This broad access to Chatwork functionalities, if misused, could lead to unauthorized data manipulation, communication, or exfiltration within Chatwork. 1. Implement granular permission controls for the `RUBE_MULTI_EXECUTE_TOOL` to restrict which specific Chatwork operations can be performed by the skill. 2. Require explicit user confirmation for sensitive Chatwork operations before execution. 3. Ensure the LLM's prompt engineering prevents it from being coerced into executing unintended or malicious Chatwork operations. | LLM | SKILL.md:55 |
Scan History
Embed Code
[](https://skillshield.io/report/8e9ea0dd1c6f7b5b)
Powered by SkillShield