ClawHub
Back to skill

Security audit

mindgraph

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly coherent for using MindGraph, but it explicitly suggests pasting API credentials into chat as a fallback, which is unsafe secret handling.

Use this skill only if you configure MINDGRAPH_TOKEN and MINDGRAPH_ACCOUNT through OpenClaw or your host's environment/secret settings. Do not paste the token into chat; if you already did, regenerate the MindGraph token. The skill otherwise appears focused on MindGraph diagram generation and edits.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

High
Confidence
98% confidence
Finding
The README explicitly tells users they can place `MINDGRAPH_BASE_URL`, `MINDGRAPH_ACCOUNT`, and `MINDGRAPH_TOKEN` directly into the chat as a fallback configuration method. Even though it says the model should avoid echoing the token, chat input is still exposed to the LLM, may be logged by the host product, retained in transcripts, or surfaced in debugging/telemetry, making credential disclosure a realistic risk.

Ssd 3

High
Confidence
98% confidence
Finding
Encouraging users to send authentication material through conversational channels normalizes insecure secret handling and increases the chance of accidental leakage. In the context of an agent skill, this is especially dangerous because the skill is designed to consume the values operationally, so secrets may traverse prompts, tool traces, logs, and third-party infrastructure before reaching the API.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.