ClawHub

Feishu Messaging

Security checks across malware telemetry and agentic risk

Overview

This is a Feishu messaging guide with visible examples for sending messages and uploading attachments, but users should be careful about recipients and files.

Install only if you intend an agent to use your Feishu bot. Use least-privilege Feishu scopes, keep app secrets out of logs and shared chats, and confirm the recipient ID, message content, and exact file path before sending or uploading.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The manifest presents the skill as a Feishu messaging workflow, but the body documents broader capabilities including document/resource management. This mismatch can mislead reviewers and users about what the skill can do, reducing informed consent and making privilege expansion harder to notice.

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill includes image and file upload workflows that are not declared in the manifest focused on sending messages. Undeclared upload behavior increases the risk of unexpected exfiltration of local/user-provided files to external Feishu APIs.

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill describes sending messages and handling media uploads without clearly warning that message contents and files will be transmitted to Feishu services. Missing disclosure can cause users to share sensitive content without understanding the external data flow.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The usage examples instruct sending messages to recipients with no warning to verify recipient identity, group ID, or sensitivity of the content. In a messaging skill, this raises the chance of misdelivery and unintended disclosure to external parties.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal