Security audit

da-fu-da-gui

Security checks for vulnerabilities and agentic risk

Overview

The skill is not overtly malicious, but it would monitor a mailbox and automatically send email replies in the background with limited documented controls.

Install only if you are comfortable granting mailbox access to a background auto-reply workflow. Before enabling it, confirm the mailbox scope, disable switch, logs, reply limits, recipient rules, and whether replies require review or are sent automatically.

A.I.G

Vulnerability Patterns

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
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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 (5)

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill explicitly states that after saving settings it will automatically begin working in the background, but the documentation does not present this behavior as a prominent warning about autonomous outbound email actions. Because the skill is granted mailbox access and can send replies without per-message confirmation, users may enable it without fully understanding the privacy, reputational, and operational consequences.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The reply template is written entirely in Chinese and provides no indication that language selection is based on user preference or a documented region-specific requirement. This creates a natural-language policy concern because it implicitly forces a locale on all users without opt-in.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The trigger phrase '自动回复，收到邮件' ('auto-reply, received email') is broad and likely to match ordinary email-related conversation, which can cause the skill to activate unintentionally. Unintended activation can make the agent send or prepare responses in the wrong context, creating privacy, integrity, or workflow risks even though the file contains only trigger text.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The file content is fully Chinese-language and does not indicate that the skill is intended only for a Chinese-speaking or region-specific audience, nor does it offer alternative language support. Under the stated policy, forcing a specific language without user opt-in can be a natural-language policy violation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 84% confidence
Finding: The trigger content is only in Chinese, which effectively constrains activation to a specific language/locale without any indication of user opt-in or fallback handling. While not directly exploitable in the same way as code execution, it can cause unexpected behavior, accessibility issues, and accidental exclusion or misrouting in multilingual environments.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.