Security Audit
taxjar-automation
github.com/ComposioHQ/awesome-claude-skillsTrust Assessment
taxjar-automation received a trust score of 83/100, 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 Potential for excessive permissions via RUBE_REMOTE_WORKBENCH, Dynamic loading of tool schemas introduces supply chain risk.
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 | Potential for excessive permissions via RUBE_REMOTE_WORKBENCH The skill `taxjar-automation` is designed for Taxjar operations. However, it exposes `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` for 'Bulk ops'. If `run_composio_tool()` can execute arbitrary Composio tools beyond the Taxjar toolkit, this could allow the agent to perform actions outside the intended scope of Taxjar automation, leading to excessive permissions. The term 'workbench' often implies a broader, less constrained execution environment. Clarify or restrict the scope of `RUBE_REMOTE_WORKBENCH` and `run_composio_tool()` to only Taxjar-related operations. If it is intended to be broader, the skill's name and description should reflect this broader capability. | LLM | SKILL.md:60 | |
| MEDIUM | Dynamic loading of tool schemas introduces supply chain risk The skill explicitly instructs the LLM to 'Always call `RUBE_SEARCH_TOOLS` first to get current tool schemas' and states 'Tool schemas change. Never hardcode tool slugs or arguments'. This means the actual behavior and available tools are dynamically fetched from the Rube MCP at runtime. While this provides flexibility, it introduces a supply chain risk: if the Rube MCP or the Taxjar toolkit definitions it serves were compromised, the skill's effective capabilities could be altered maliciously without any change to the skill's static definition. Implement mechanisms to verify the integrity and authenticity of dynamically loaded tool schemas (e.g., cryptographic signatures). Consider pinning to specific versions of tool schemas where possible, or at least providing clear versioning information. | LLM | SKILL.md:18 |
Scan History
Embed Code
[](https://skillshield.io/report/e104430fad271ae9)
Powered by SkillShield