ClawHub
Back to skill

Security audit

Auto Deep Research

Security checks across malware telemetry and agentic risk

Overview

This deep-research skill appears to do what it says: search the web, read pages, and save notes and reports, with privacy cautions but no hidden or destructive behavior found.

Install only if you are comfortable with research queries and selected URLs being sent to external search or page-reading services. Use limited-scope API keys, keep .env files out of version control, avoid private or token-bearing URLs, and delete output folders when saved research notes are no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (9)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill clearly directs use of shell scripts and curl-based commands, but it does not declare permissions or otherwise signal that command execution is required. That mismatch can cause an agent or user to invoke shell/network-capable behavior without informed consent or proper sandboxing assumptions.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs creation of multiple files under output/{topic-slug}/ and ongoing writes to state, memo, sources, and report files, but gives no warning that it will modify the filesystem. Silent filesystem writes are risky because they can overwrite existing data, expose sensitive research content locally, or surprise users in restricted environments.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill directs sending search queries and page URLs to third-party services such as Tavily, DuckDuckGo, Jina, or browser/web fetch tooling without warning that user input and targets will leave the local environment. This is dangerous because research topics may contain confidential prompts, internal URLs, or sensitive investigative context that gets disclosed externally.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill tells users to place API keys in environment variables or a .env file but provides no guidance on secure storage, scope limitation, or avoiding accidental disclosure. This increases the chance of secrets being committed, logged, inherited by subprocesses, or exposed to unrelated tools executed by the agent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script transmits any supplied URL to the external service r.jina.ai, which means sensitive internal URLs, private endpoints, or query parameters could be disclosed to a third party. In a deep-research skill, users may reasonably pass arbitrary links, including confidential or internal resources, so this creates a real privacy and data-exposure risk even though the functionality appears intentional.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The script sends the raw user query to third-party search providers (Tavily or DuckDuckGo) without any disclosure, consent, or guardrails. In a research skill, user queries may contain sensitive internal topics, personal data, or confidential investigation targets, so external transmission creates a real privacy and data-handling risk even if this is functionally expected.

External Transmission

Medium
Category
Data Exfiltration
Content
**手动调用**：
```bash
# Tavily（需要 API key）
curl "https://api.tavily.com/search" \
  -H "Authorization: Bearer $TAVILY_API_KEY" \
  -d '{"query": "query", "max_results": 5}'
Confidence
97% confidence
Finding
curl "https://api.tavily.com/search" \ -H "Authorization: Bearer $TAVILY_API_KEY" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
**手动调用**：
```bash
# Tavily（需要 API key）
curl "https://api.tavily.com/search" \
  -H "Authorization: Bearer $TAVILY_API_KEY" \
  -d '{"query": "query", "max_results": 5}'
Confidence
97% confidence
Finding
https://api.tavily.com/

External Transmission

Medium
Category
Data Exfiltration
Content
-d '{"query": "query", "max_results": 5}'

# DuckDuckGo（免费，搜索网页内容）
curl "https://api.duckduckgo.com/?q=query&format=json"
# 如果curl失效，则用 Python duckduckgo-search 库或 SearXNG
```
Confidence
95% confidence
Finding
https://api.duckduckgo.com/

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.