Security Audit
twocaptcha-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
twocaptcha-automation received a trust score of 85/100, 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, 0 high, 1 medium, and 0 low severity. Key findings include Generic Tool Execution via Remote Workbench.
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 20, 2026 (commit 27904475). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Generic Tool Execution via Remote Workbench The skill describes the `RUBE_REMOTE_WORKBENCH` tool for 'Bulk ops' using `run_composio_tool()`. This suggests a generic execution environment capable of invoking arbitrary Composio tools. Without explicit constraints on the types of tools or operations allowed within this workbench, or proper sandboxing, this capability could lead to excessive permissions being exercised by the LLM if it's prompted to execute tools with broad access (e.g., filesystem, network, or sensitive APIs) that are not directly related to Twocaptcha. The skill itself does not define these constraints, making it a potential vector for an LLM to invoke powerful, unconstrained operations if not properly managed by the host system. Clarify the scope and limitations of `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()` within the skill description. Specify if it's sandboxed, what types of tools it can execute, and what permissions those tools have. Ensure the host environment strictly limits the capabilities of the workbench and the LLM's access to it. | LLM | SKILL.md:50 |
Scan History
Embed Code
[](https://skillshield.io/report/c1d4fa5ff000de07)
Powered by SkillShield