Back to skill

Security audit

明歌内训课内容加工

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Feishu course-review workflow that reads a user-provided document and creates four derived Feishu documents, with no hidden code, persistence, or unrelated behavior found.

Before installing, understand that this skill will use Feishu document access to read course materials and create four new child documents containing derived content. Use it only on documents you are authorized to process and share, especially if they contain participant names, comments, or private training transcripts.

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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The documented trigger phrases are broad and overlap with common natural-language requests such as '帮我总结' or '帮我复盘', which can cause the skill to activate in contexts the user did not intend. Because activation also expects a Feishu document link and then performs multi-step processing and document creation, accidental triggering could lead to unintended access, processing, or generation actions against user-provided documents.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill processes potentially sensitive course materials, participant identities, timestamps, and comments from a Feishu document, then derives and reproduces that content into new documents. Without any user-facing privacy notice, consent checkpoint, or data-handling warning, users may unknowingly expose personal or confidential information through secondary documents and broader document propagation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The workflow instructs the agent to create four remote Feishu wiki child documents and write generated content into them automatically. Because there is no explicit warning or confirmation that remote documents will be created and modified, a user could trigger unintended writes, causing confidential data duplication, workspace clutter, or unauthorized changes in a shared knowledge base.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

该技能文件从标题到正文均仅以中文表述,没有说明这是面向特定中文用户群、特定区域场景,或提供其他语言选项。根据规则,未获用户选择或未给出明确地域/合规理由的单一语言约束,属于自然语言层面的语言/locale 政策风险。

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The activation phrases, examples, and operating instructions all assume Chinese-language invocation and output behavior, but the file does not explicitly offer users a language option or justify a Chinese-only constraint. Under the policy, forcing a specific language without opt-in can be a natural-language policy issue.

Content

No source excerpt is available for this finding.