Security Audit

paywall-upgrade-cro

github.com/coreyhaines31/marketingskills

AI SkillCommit a04cb61a577a

TRUSTED

Scanned 4 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

paywall-upgrade-cro 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 Potential Data Exfiltration via Local File Read.

The analysis covered 4 layers: dependency_graph, manifest_analysis, static_code_analysis, llm_behavioral_safety. All layers scored 70 or above, reflecting consistent security practices.

Last analyzed on February 16, 2026 (commit a04cb61a). 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

Excessive Permissions

1 finding

Security Findings1

Severity	Finding	Layer	Location
HIGH	Potential Data Exfiltration via Local File Read The skill explicitly instructs the LLM to read the local file `.claude/product-marketing-context.md` and use its content as context for generating responses. This grants the LLM direct access to local filesystem content. If this file contains sensitive, proprietary, or confidential product marketing information (e.g., unreleased features, competitive analysis, internal strategies), the LLM could inadvertently expose this data to the user or an external system through its generated output. This also represents an excessive permission granted to the LLM. 1. Restrict File Content: Ensure that `.claude/product-marketing-context.md` contains only non-sensitive, publicly shareable information. 2. Sanitize Input: If sensitive data is necessary, pre-process the file to remove or redact sensitive portions before it is provided to the LLM. 3. Remove Direct Access: Re-evaluate if the LLM truly needs direct read access to a local file. Consider embedding necessary context directly into the skill's prompt or using a secure, controlled data retrieval mechanism. 4. Output Filtering: Implement strict output filtering and monitoring to prevent the LLM from disclosing sensitive information.	Unknown	SKILL.md:10

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/88d9ac36c9ef904a.svg)](https://skillshield.io/report/88d9ac36c9ef904a)