ClawHub

CueCue Deep Research

Security checks across malware telemetry and agentic risk

Overview

This is a coherent financial research skill that uses an external CueCue CLI and API key, with expected privacy and storage cautions but no artifact-backed malicious behavior.

Install only if you are comfortable trusting the @sensedealai/cuecue npm package and sending financial research prompts to CueCue. Use a dedicated API key, avoid submitting confidential client or nonpublic business information unless approved, choose report output paths intentionally, and use foreground mode when you want closer supervision of long-running research.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The top-level description says the skill should be proactively invoked for essentially any request needing data-backed financial analysis, without clear gating criteria, consent boundaries, or narrower scope. In an agent setting this can cause unintended tool use, unnecessary transmission of user prompts to an external service, and over-delegation on ordinary financial questions that may not require the tool.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The instruction '凡涉及以下金融研究需求，均应主动调用本技能' directs the agent to use the skill for a very broad class of financial topics with no limiting conditions. This increases the chance of unnecessary external calls, surprise file creation, and unintended disclosure of user-supplied financial queries or sensitive business context.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation instructs the agent to run a networked CLI that sends research queries to an external service and writes reports to disk, but it does not warn about data leaving the local environment or about local artifact creation. In practice this can expose confidential prompts, company strategy, or investment research topics to a third party and leave sensitive reports persisted on the host unexpectedly.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal