Security Audit

missive-automation

github.com/ComposioHQ/awesome-claude-skills

AI SkillCommit 27904475d127

TRUSTED

Scanned about 2 months ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

missive-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 2 findings: 0 critical, 1 high, 0 medium, and 0 low severity. Key findings include Broad Tool Execution via Rube MCP, Unpinned Rube MCP Dependency.

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

Manifest Analysis

100%

Static Code Analysis

100%

Dependency Graph

100%

LLM Behavioral Safety

85%

Behavioral Risk Signals

Network Access

1 finding

Filesystem Write

1 finding

Shell Execution

1 finding

Excessive Permissions

1 finding

Security Findings2

Severity	Finding	Layer	Location
HIGH	Broad Tool Execution via Rube MCP The skill instructs the LLM to use `RUBE_SEARCH_TOOLS` to discover available tools and then `RUBE_MULTI_EXECUTE_TOOL` or `RUBE_REMOTE_WORKBENCH` to execute them. While the skill's stated purpose is Missive automation, the Rube MCP itself is a general-purpose tool execution platform. There are no explicit constraints in the skill's instructions or manifest to limit the LLM to only Missive-related tools. This means an LLM could be prompted to discover and execute any tool exposed by the Rube MCP, potentially leading to unintended actions, data exfiltration, or system compromise if the MCP exposes privileged tools (e.g., file system access, network requests, shell commands). The mention of `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` in the 'Quick Reference' table is particularly concerning as it suggests a generic tool execution capability. 1. Restrict `RUBE_SEARCH_TOOLS`: Modify the skill to explicitly filter or limit the scope of tools that `RUBE_SEARCH_TOOLS` can return to only those relevant to Missive. 2. Explicitly whitelist tools: If possible, configure the Rube MCP or the skill's environment to only allow execution of a predefined whitelist of Missive tools. 3. Avoid generic execution: Remove or restrict the use of `RUBE_REMOTE_WORKBENCH` with `run_composio_tool()` if its capabilities are too broad for the skill's intended purpose. 4. Least Privilege: Ensure the Rube MCP connection itself operates with the least necessary privileges.	LLM	SKILL.md:40
INFO	Unpinned Rube MCP Dependency The skill manifest specifies a dependency on `rube` via the `mcp` field (`"mcp": ["rube"]`) but does not pin a specific version. While common for MCPs, this means that updates to the Rube MCP could potentially introduce breaking changes, new vulnerabilities, or altered tool behaviors without explicit review. If possible and supported by the Rube MCP, specify a version range or exact version for the `rube` dependency in the manifest to ensure consistent behavior and allow for controlled updates.	LLM	SKILL.md:3

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/1c800e817ece0fed.svg)](https://skillshield.io/report/1c800e817ece0fed)