ClawHub
Back to skill

Security audit

Chart AI

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches its ChartGen purpose, but it sends user data to a third-party API and lets API responses drive unscoped local downloads/writes, so users should review it carefully before installing.

Install only if you are comfortable sending selected prompts and spreadsheet files to ChartGen. Use a revocable ChartGen API key, avoid confidential or regulated data unless your policy allows it, and be aware that API-returned artifacts are downloaded and saved locally.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (5)

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The tool trusts a server-provided download_url and fetches it with no host allowlist, scheme restriction beyond http/https, size limits, or redirect validation. A compromised or malicious API response could coerce the local agent into making arbitrary outbound requests and writing attacker-controlled content to disk, creating an SSRF-style primitive and untrusted file write behavior.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation description is broad enough to trigger on many common requests involving files, reports, spreadsheets, or visualizations, which can route users into an external API workflow more often than necessary. Over-broad invocation increases the chance of unnecessary third-party data transfer, including sensitive spreadsheet contents, without a narrowly scoped user intent match.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
Local file contents are uploaded to a third-party service without any explicit consent gate or user-facing disclosure at the transmission point. In an agent setting, users may assume uploaded spreadsheets are processed locally, so silent transfer can expose sensitive business or personal data to an external provider.

Missing User Warnings

Medium
Confidence
78% confidence
Finding
The user's query and metadata are sent to the remote ChartGen API without an explicit warning or confirmation. Prompts often contain confidential business context, and in this skill context the external transmission is central but still should be clearly disclosed to avoid inadvertent data leakage.

Unrestricted Tool Access

Medium
Category
Excessive Agency
Content
- With files: "ChartGen is analyzing your data, ~2–5 min..."
- PPT: "ChartGen is generating your PPT, ~10–20 min, please be patient..."

**Then call the tool:**
```
node tools/chartgen_api.js submit "<query>" <channel> [files...]
```
Confidence
91% confidence
Finding
tool:*

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.