ClawHub
Back to skill

Security audit

LinkMind Interaction

Security checks for vulnerabilities and agentic risk

Overview

This skill is purpose-aligned for managing LinkMind channels, but it needs review because it can use ambient API keys and perform account-changing actions including posts and channel deletion.

Install only if you trust this publisher and intend to let an agent manage your LinkMind account. Use an explicit --config path in a private LinkMind-specific key file, avoid relying on .env or generic config files, and require explicit confirmation before sending messages, disabling channels, or deleting channels.

SkillSpector

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
87% confidence
Finding
The invocation description is broad enough to trigger on many generic requests to manage channels or messages, while the skill includes write and destructive actions such as create, disable, and delete. Over-broad routing can cause the agent to invoke this skill in situations where the user did not intend remote account changes, increasing the risk of unintended subscriptions, message posting, or channel deletion.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill documents destructive deletion of channels and only shows a CLI-level --yes flag, but it does not require an explicit interactive confirmation or warn the user at the skill-policy level before performing deletion. In an agent setting, this can translate into irreversible remote data loss if the model misinterprets a request, selects the wrong channel, or executes a deletion too eagerly.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The script automatically searches environment variables and multiple local files, including generic names like config.md and .env, to extract an API key without any explicit user confirmation or prominent disclosure. In an agent/skill context, this creates a real secret-harvesting risk because the tool can silently consume credentials present in the working directory or environment and use them to act on the user's hosted LinkMind account.

Static analysis

No suspicious patterns detected.