Back to skill

Security audit

Auto Contents

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed automation guide for a local content workflow, with real posting and preference-memory risks that are mostly purpose-aligned and bounded.

Install this only if you want an agent to operate your configured MakeContents, WeChat, Feishu, and optionally Xiaohongshu workflow. Review the account/API-key configuration first, keep Xiaohongshu publishing disabled unless needed, and periodically inspect or clear the learned preference file if you do not want long-term behavioral profiling retained.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill advertises itself for a very broad set of loosely scoped "content" tasks, including fetching, filtering, creating, publishing, and archiving, with little activation boundary. In an agent setting, this can cause over-invocation and unintended autonomous execution of sensitive actions such as publishing or pushing content when a user's request is only partially related.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill hardcodes Chinese-language and specific platform style requirements (for example, Xiaohongshu-style content) without indicating that this depends on user preference. This can override user intent, cause inappropriate output routing or formatting, and increase the chance an agent produces or publishes content in an undesired language/style.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The document explicitly says the agent will automatically maintain a memory document derived from user history and preserve historical records, but it provides no notice, consent flow, retention limit, or data-minimization guidance. In a content automation skill that profiles user selection and publishing preferences, this creates a privacy and profiling risk because behavioral data may be stored persistently without the user's informed awareness.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The file is written as an API reference explicitly for agent use and presents all operational instructions in Chinese, with no indication that other languages are supported or that the locale is user-selectable. Under the policy, forcing a specific language without opt-in is a natural-language policy violation unless clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The API allows an agent to publish content to Xiaohongshu, which is a real-world external side effect with reputational and compliance risk. Although the document includes a mitigation requiring agents to set is_private=true and respect the xhs_enabled switch, the endpoint itself still exposes publication capability and would be unsafe if an agent ignored or mishandled those constraints.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

All user-facing text in the file is Chinese, and there is no statement offering a language or locale choice. If this skill is intended for general use, hard-coding a single language without opt-in can violate language-choice policy expectations.

Content

No source excerpt is available for this finding.