Security Audit

affaan-m/everything-claude-code:docs/ja-JP/skills/eval-harness

github.com/affaan-m/everything-claude-code

AI SkillCommit 9a478ad67605

CRITICAL

Scanned 18 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

affaan-m/everything-claude-code:docs/ja-JP/skills/eval-harness received a trust score of 25/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.

SkillShield's automated analysis identified 2 findings: 1 critical, 1 high, 0 medium, and 0 low severity. Key findings include Potential Command Injection via User-Controlled Input, Excessive Permissions: Broad Filesystem and Shell Access.

The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Static Code Analysis layer scored lowest at 55/100, indicating areas for improvement.

Last analyzed on March 20, 2026 (commit 9a478ad6). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

100%

Static Code Analysis

55%

Dependency Graph

100%

LLM Behavioral Safety

100%

Behavioral Risk Signals

Network Access

1 finding

Filesystem Write

1 finding

Shell Execution

2 findings

Dynamic Code

2 findings

Excessive Permissions

2 findings

Security Findings2

Severity	Finding	Layer	Location
CRITICAL	Potential Command Injection via User-Controlled Input The skill declares access to the 'Bash', 'Grep', and 'Glob' tools, indicating its ability to execute shell commands. The integration patterns `/eval define <feature-name>`, `/eval check <feature-name>`, and `/eval report <feature-name>` suggest that the `<feature-name>` argument is user-controlled input. If this input is directly interpolated into shell commands without proper sanitization or escaping, it could lead to command injection. An attacker could provide a `feature-name` like `my-feature; rm -rf /` to execute arbitrary commands on the host system. All user-controlled input passed to shell commands must be rigorously sanitized and escaped. Consider using a dedicated library for command execution that handles argument escaping, or strictly validate input against an allow-list of characters and patterns. Avoid direct string concatenation for constructing shell commands.	Static	SKILL.md:100
HIGH	Excessive Permissions: Broad Filesystem and Shell Access The skill requests broad permissions including 'Read', 'Write', 'Edit' for filesystem access, and 'Bash', 'Grep', 'Glob' for shell execution. While the skill's purpose (running evaluations, tests, builds) might necessitate some of these, the combination of 'Write' and 'Bash' with user-controlled input (as identified in the command injection finding) creates a significant attack surface. The filesystem access is not scoped to a specific directory, allowing potential modification or reading of sensitive files outside the intended scope if combined with a command injection vulnerability. Review and restrict tool access to the absolute minimum required functionality. If filesystem access is needed, scope it to a specific, isolated directory. Implement robust input validation and sanitization for all user-provided arguments used in conjunction with these powerful tools to prevent their misuse.	Static	SKILL.md

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/73bea14f7cb3a54c.svg)](https://skillshield.io/report/73bea14f7cb3a54c)