Security Audit

templated-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 99e2a2951545

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

templated-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 Capabilities.

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

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Network Access

1 finding

Shell Execution

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Broad Tool Execution Capabilities The skill provides the LLM agent with broad capabilities to discover and execute arbitrary tools from the 'templated' toolkit and potentially other Composio tools via the Rube MCP. Specifically, `RUBE_MULTI_EXECUTE_TOOL` allows execution of any discovered tool with arbitrary arguments, and `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` suggests an even broader ability to execute Composio tools. This means a malicious prompt could instruct the agent to execute sensitive or destructive operations if such tools are exposed by the underlying MCP. The skill itself does not implement any restrictions on which tools can be executed or what arguments can be passed, relying entirely on the security of the external MCP and its exposed toolkits. Implement stricter controls or allowlisting for tool execution. Instead of allowing arbitrary tool discovery and execution, consider defining a specific set of safe tools and operations that the agent is permitted to perform. If `RUBE_REMOTE_WORKBENCH` is not strictly necessary, consider removing its mention or restricting its use. Ensure that the `templated` toolkit itself only exposes safe and necessary operations.	LLM	SKILL.md:57

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/5dbf7d960b4a1bf4.svg)](https://skillshield.io/report/5dbf7d960b4a1bf4)