Back to skill

Security audit

飞书+Agent企业方案

Security checks for vulnerabilities and agentic risk

Overview

This is a small, non-executable Lark/Feishu enterprise automation skill whose broad trigger words merit care but whose artifacts show no hidden, persistent, or destructive behavior.

Install this only if you want a Chinese-language Feishu/Lark enterprise automation skill. Because the advertised use cases include document, table, calendar, and task changes, confirm sensitive or state-changing actions before letting an agent use connected Feishu credentials or business data.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases are generic business terms such as '企业知识库', '飞书自动化', and '新人培训', which are likely to appear in normal enterprise conversations. This can cause unintended skill activation, leading the agent to invoke enterprise automation behaviors in contexts where the user did not explicitly request this skill, increasing the chance of accidental data access or workflow actions.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The trigger list is very broad and generic for an enterprise-focused skill, which increases the chance of accidental or overly permissive activation in unrelated contexts. In an enterprise automation and knowledge-base setting, unintended activation can expose internal workflows, cause the agent to handle sensitive business queries without clear user intent, or create opportunities for prompt-routing abuse.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The natural-language description is presented only in Chinese and does not state that the skill is China-specific or offer any language/locale choice. This can indicate a language policy issue where a specific language is imposed without opt-in or justification.

Content

No source excerpt is available for this finding.