Security audit

微信Mac版消息发送

Security checks for vulnerabilities and agentic risk

Overview

This skill is meant to send WeChat messages, but it can send immediately to an unverified recipient and builds unsafe automation commands from user-provided text.

Install only if you are comfortable with an agent controlling the WeChat Mac UI and sending messages through your logged-in account. Before use, the skill should add a visible confirmation step showing the resolved contact and exact message, require exact recipient verification, avoid partial-name top-result selection, and pass message/contact text as data rather than interpolating it into shell or AppleScript source.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (2)

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:34
Finding: Command and AppleScript Injection Through Unsanitized Placeholder Substitution
Content: View full analysis
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:40
Finding: Messages Can Be Sent to an Unverified or Incorrect Recipient
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Missing User Warnings

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The skill automates WeChat by selecting the top search result, typing message content, and pressing Enter to send immediately, but it does not clearly warn that it will transmit a real message without human review. This is especially dangerous because the workflow relies on fragile coordinates and implicit focus assumptions, so mis-selection of the contact or mistyped content can directly cause unintended outbound communication.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrases are broad, everyday expressions such as '发微信' and '给 XXX 发微信', which can plausibly appear in normal conversation and cause unintended activation. Because this skill performs direct UI automation to send a real message without an explicit confirmation step, accidental triggering can lead to messages being sent to the wrong person or at the wrong time.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.