Security audit

ME Event 精选

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed event-notification helper that uses fixed read-only ME News API endpoints and local state, with user confirmation required before recurring reminders are created.

Install this if you want ME News event browsing or recurring AI/Web3 event reminders. Before enabling subscription mode, confirm the event filters, delivery channel, and recurring schedule; do not place channel credentials in the skill directory or state file.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The default prompt broadly instructs the agent to use the skill to configure filters and check featured events without clear user-trigger boundaries or narrower activation conditions. This can cause the skill to be invoked in situations where the user did not explicitly request subscription setup or event monitoring, increasing the risk of overreach and unintended actions.

Natural-Language Policy Violations

Medium

Confidence: 93% confidence
Finding: The skill metadata is written to force Chinese-language presentation and prompting, but there is no indication that the user requested Chinese. This can mislead users, reduce transparency about what the skill will do, and impair informed consent or review of actions if the user's preferred language is different.

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/args.test.mjs:9

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/delivery.test.mjs:12

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/initialization.test.mjs:34