Security Audit

ailabs-393/ai-labs-claude-skills:packages/skills/research-paper-writer

github.com/ailabs-393/ai-labs-claude-skills

AI SkillCommit 1a12bc7aadcc

TRUSTED

Scanned 18 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

ailabs-393/ai-labs-claude-skills:packages/skills/research-paper-writer 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 1 finding: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Reflected Input in Skill Output Leading to Potential Prompt Injection.

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 March 14, 2026 (commit 1a12bc7a). 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

93%

Behavioral Risk Signals

Shell Execution

1 finding

Dynamic Code

1 finding

Security Findings1

Severity	Finding	Layer	Location
MEDIUM	Reflected Input in Skill Output Leading to Potential Prompt Injection The `index.js` skill directly reflects the `input` it receives as part of its return message. If this `input` contains malicious instructions (e.g., 'ignore previous instructions' or other prompt injection attempts), and the host LLM processes the skill's output without proper sanitization or context separation, it could lead to prompt injection, allowing an attacker to manipulate the host LLM's subsequent behavior or instructions. Avoid directly reflecting untrusted user input in skill outputs that are fed back to the LLM. If the input must be returned, ensure it is properly sanitized, escaped, or clearly demarcated as user data (e.g., by wrapping it in XML tags or specific delimiters) so that the LLM does not interpret it as instructions. For example, instead of `input`, return `user_input: input` or `raw_user_query: input`.	LLM	index.js:6

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/a98e4c330114edc3.svg)](https://skillshield.io/report/a98e4c330114edc3)