Security Audit

sourcegraph-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

sourcegraph-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 Excessive Permissions 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 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

Filesystem Write

1 finding

Shell Execution

1 finding

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Excessive Permissions via Remote Workbench The skill exposes the `RUBE_REMOTE_WORKBENCH` tool, which, according to the documentation, can be used for 'Bulk ops' with `run_composio_tool()`. This implies the ability to execute arbitrary Composio tools in a remote environment. If the underlying Composio tools have broad permissions (e.g., filesystem access, network access, or arbitrary code execution), an LLM agent, if compromised or misdirected, could leverage this tool to perform unauthorized actions, data modification, deletion, or exfiltration beyond the intended scope of Sourcegraph automation. While the documentation advises caution, the inherent power of this tool presents a significant risk when exposed to an autonomous agent. Implement strict access controls and guardrails around the `RUBE_REMOTE_WORKBENCH` tool. Consider requiring human approval for operations involving this tool or limiting its scope to specific, pre-approved Composio tools and operations. Ensure that the `run_composio_tool()` function within Rube MCP has fine-grained permissions and robust auditing. The skill documentation should also explicitly highlight the security implications of using this powerful tool.	LLM	SKILL.md:68

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/c029b05bf14dbc86.svg)](https://skillshield.io/report/c029b05bf14dbc86)