feishu-video

• Outcome: Review
• Latest audit: May 29, 2026 at 9:24 PM UTC
• Version:
  1.0.0

Overview

This appears to be a real Feishu/Lark messaging helper, but it under-discloses video and group-chat sending while using credentials that can upload and send media externally.

Review carefully before installing. Use a least-privilege Feishu/Lark bot, keep app secrets out of shell history and logs, verify user and chat IDs before each send, and assume any selected audio or video file will be uploaded to Feishu/Lark and delivered to the chosen recipient or group.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

High

• Confidence: 99% confidence
• Finding: The file implements end-to-end video messaging, including MP4 upload, segmentation, and media-message sending, while the skill is declared as an audio/voice-message skill. This capability mismatch is dangerous because users or reviewers may grant trust, permissions, and credentials under the assumption the skill only handles audio, but it can exfiltrate or transmit video content instead.

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The code introduces extra video-processing features such as ffmpeg-based cover extraction and segmentation that are unrelated to an audio-message skill. These unjustified capabilities broaden the attack surface, enable larger or more complex media transmission workflows, and make hidden misuse harder to detect during review.

Intent-Code Divergence

High

• Confidence: 98% confidence
• Finding: The file header and usage text explicitly describe a Feishu video sender, which contradicts the declared audio-message intent of the skill. This deceptive or careless documentation mismatch increases the likelihood that risky functionality will evade review and be executed with sensitive credentials under false expectations.

Missing User Warnings

Medium

• Confidence: 75% confidence
• Finding: The README instructs users to upload and send audio/video content and configure application credentials, but it does not clearly warn that user media will be transmitted to Feishu/Lark and delivered to other users or chats. In an agent-skill context, missing disclosure about external transmission of potentially sensitive media increases the risk of unintended exfiltration or privacy violations.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: This skill instructs users to upload audio content and provide Feishu app credentials, but it does not clearly disclose that audio data and metadata will be transmitted to Feishu's servers. That omission can lead users to unintentionally expose sensitive voice content, personal data, or regulated information through a third-party service.

VirusTotal

66/66 vendors flagged this skill as clean.