Ali Feishu Sender

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:32 PM UTC
• Version:
  1.0.0

Overview

This skill is a Feishu/Lark sender that uploads selected text and media as advertised, with privacy and local file-conversion cautions but no hidden or deceptive behavior found.

Install only if you want your agent to upload selected content to Feishu/Lark. Use least-privilege Feishu credentials, confirm the destination recipient or group before sending, avoid sensitive files unless appropriate, and be aware that media conversion may leave .opus, .mp4, or _fs.mp4 files next to the originals.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger conditions are broad enough to match common requests such as 'send to feishu/lark' or posting generated content, which may cause the skill to activate in situations where the user did not intend external transmission. In the context of a messaging skill, over-broad routing is dangerous because it can lead to unintended exfiltration of text or media to a third-party service.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill description says it sends multimedia content to Feishu but does not provide a prominent user-facing warning that content will be uploaded to an external third-party service using configured credentials. This is a meaningful security/privacy weakness because users may provide sensitive images, audio, or documents without realizing they are leaving the local environment and being shared externally.

Missing User Warnings

Medium

• Confidence: 83% confidence
• Finding: The module is designed to transmit message content and uploaded files to Feishu, but it provides no built-in confirmation, disclosure, or policy guardrails before sending data off-host. In an agent setting, this increases the chance of unintended disclosure of sensitive local content because sending is one function call away and uses default credentials and recipient state.

Missing User Warnings

Medium

• Confidence: 79% confidence
• Finding: Audio conversion silently creates a derived .opus file next to the source file, which can unexpectedly persist sensitive transformed content on disk. In shared or monitored environments, this can expand data exposure and leave artifacts users did not intend to keep.

Missing User Warnings

Medium

• Confidence: 80% confidence
• Finding: Video processing silently creates or rewrites derived MP4 files, potentially leaving additional media artifacts on disk without user awareness. For agent-driven workflows, this is risky because local files may contain sensitive content and persisted derivatives can remain after the send completes.

VirusTotal

66/66 vendors flagged this skill as clean.