ClawHub

ZeeLin Deep Research 深度研究

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed ZeeLin research API connector that sends prompts to ZeeLin, retrieves generated reports, and delivers them back to the user.

Install only if you are comfortable sending research prompts and generated reports to ZeeLin, and on Feishu, placing report content into a Feishu document. Avoid submitting secrets, regulated data, or confidential business material unless ZeeLin and the delivery platform are approved for that data, and delete temporary report files after delivery when handling sensitive content.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
83% confidence
Finding
The skill goes beyond simple research generation and instructs the agent to create Feishu documents, write extracted report contents into them, and send links. This expands the action surface from passive API use to cross-system content publication, which can expose sensitive research results to third-party platforms or the wrong recipient if context, tenant, or permissions are mis-scoped.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill sends user prompts and session identifiers to an external service without any privacy notice, consent language, or data-handling disclosure. Research prompts may contain proprietary, personal, or regulated information, so silent transmission to a third-party API creates a real confidentiality and compliance risk.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill instructs the agent to download generated reports to local temporary storage and then forward them, but gives no warning about local persistence, cleanup, or access controls. This can leave sensitive reports on disk or in shared temporary paths, increasing the chance of unintended disclosure on multi-tenant or reused environments.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal