全球AI新闻简报

• Outcome: Pass
• Latest audit: Aug 20, 2026 · 8:03 AM UTC
• Version: 1.0.0

Overview

This skill does what it claims: it searches public social/news platforms through RedFox and creates local news-intelligence reports, with no hidden destructive or unrelated behavior found.

Install only if you are comfortable sending search keywords and receiving aggregated platform data through RedFox using your REDFOX_API_KEY. Avoid using sensitive internal investigation terms unless your organization accepts that third-party API exposure, and consider limiting searches to a platform subset when appropriate.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 92% confidence
• Finding: The skill instructs the agent to read an API key from environment variables, send user-supplied queries and collected content to an external host, and write an HTML report to disk, but it does not declare any permissions. This creates a transparency and governance gap: an agent or reviewer cannot accurately assess that the skill performs network access, local file writes, and secret consumption before it runs.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The skill is designed to trigger from broad natural-language phrases such as '新闻情报', '全网搜索', and 'global news' without clear boundaries on when it should activate. In an agent environment, this can cause the skill to run unintentionally on loosely related requests and send user queries to external services or initiate multi-platform collection without explicit user consent for that action.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The README mentions the need for a third-party API key but does not clearly warn users that their query content will be transmitted to an external provider and used to aggregate data from 11 platforms. This creates a transparency and privacy risk because users may disclose sensitive topics, names, or internal terms without understanding that the data leaves the local agent context.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger phrases are extremely broad, covering generic requests such as news search, analysis, tracking, and industry intelligence. Broad triggers increase the chance that the skill activates in contexts where the user did not intend external transmission of their query or mass cross-platform collection, causing unnecessary data exposure and overreach.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill sends user search terms and retrieves cross-platform content through an external service at redfox.hk, but the description does not clearly warn users that their queries and resulting data are transmitted off-platform. In a news-intelligence context, queries may contain sensitive topics, internal investigations, or person/company names, so lack of disclosure raises privacy, confidentiality, and compliance risk.

VirusTotal

61/61 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.