Security audit

Thought Mining

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only Chinese writing-assistant skill; its broad triggers and web fact-checking deserve user awareness but do not show hidden or harmful behavior.

Installers should know this skill may guide ordinary writing-help requests into a multi-step brainstorming workflow and may use web search for fact checking. Avoid sharing sensitive, unpublished, or personal draft details in search-backed validation unless you are comfortable with that external lookup.

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 (4)

Vague Triggers

Medium

Confidence: 96% confidence
Finding: The trigger phrases include broad everyday requests like '帮我整理想法' and '写不出来', which can match ordinary conversation and cause the skill to activate unintentionally. This increases the chance that users are routed into a specialized workflow without clear consent, potentially altering handling of sensitive writing-related content.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The skill instructs the agent to use web_search to validate user ideas and facts, but it does not warn that user-provided content, claims, or draft details may be transmitted to external services. In a brainstorming and writing context, users may share unpublished, personal, or sensitive material, so silent external querying creates a meaningful privacy and confidentiality risk.

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The usage section repeats vague startup keywords without defining when the skill should or should not engage. Because these phrases are common in normal writing assistance requests, the skill may over-trigger and capture user ideation or draft content in contexts where the user did not intend this structured process.

Natural-Language Policy Violations

Low

Confidence: 78% confidence
Finding: This markdown file contains end-user instructional content exclusively in Chinese, and there is no indication that the user can choose another language or that the locale restriction is required for a specific regional purpose. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.