Security Audit

amara-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

amara-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 Skill exposes broad tool execution capabilities 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

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Shell Execution

1 finding

Dynamic Code

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Skill exposes broad tool execution capabilities via Rube MCP The skill documentation explicitly mentions and encourages the use of `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` for 'Bulk ops'. This function, when exposed to an LLM agent, allows the agent to execute arbitrary Composio tools accessible through the Rube MCP, not just those specifically related to Amara. This grants very broad and potentially excessive permissions to the agent. If the agent is compromised (e.g., via prompt injection), an attacker could leverage this capability to perform actions across various integrated services, potentially leading to data exfiltration, modification, or deletion beyond the intended scope of the Amara skill. 1. Principle of Least Privilege: If `RUBE_REMOTE_WORKBENCH` is not strictly necessary for the core Amara automation tasks, consider removing its mention or restricting its use within the skill's intended workflow. 2. Contextual Guardrails: Implement strong guardrails around the `RUBE_REMOTE_WORKBENCH` tool call within the agent's orchestration logic to ensure it's only used for explicitly approved and limited operations. 3. Tool-specific Permissions: If possible, configure Rube MCP to limit the scope of `run_composio_tool()` to only specific toolkits or actions relevant to the Amara skill, rather than all available Composio tools. 4. User Confirmation: For high-impact operations initiated via `RUBE_REMOTE_WORKBENCH`, require explicit user confirmation before execution.	LLM	SKILL.md:60

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/ae4b2a0209038566.svg)](https://skillshield.io/report/ae4b2a0209038566)