Security audit

Beatclaw

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly aligned with generating and selling beats, but it needs Review because it collects API keys through chat and can overwrite its installed instructions from a mutable remote URL.

Review this skill before installing. Use it only if you trust BeatClaw and its remote update URL, inspect downloaded updates before allowing them to replace SKILL.md, and prefer a secure dashboard or secret store for API keys instead of pasting reusable credentials into chat. Confirm each generation, stem split, listing update, and deletion because the agent can spend third-party credits and change public marketplace listings.

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 (6)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The setup instructs users to download remote content directly into the agent's skills directory with curl, but provides no integrity verification, review step, pinning, or warning about the trust implications. This creates a supply-chain risk: if the remote endpoint or transport path is compromised, a malicious skill could be installed and automatically loaded by the agent in future sessions.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: Telling the agent to install the skill from a URL causes the agent to fetch remote content and save files automatically, yet the instructions omit any warning that this delegates network access and filesystem writes to the agent. In an agent context, this is especially risky because users may not realize they are authorizing installation of unreviewed instructions that will persist and execute as part of future sessions.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The setup states that the agent will handle API key storage and configuration automatically for owner email, PayPal email, and Suno API credentials, but it does not explain where secrets are stored, how they are protected, or what data may be transmitted to third parties. Because the skill is designed to collect payment and API credentials, the omission is more dangerous in context: users are encouraged to provide sensitive data to an auto-configuring agent without informed consent or storage safeguards.

Ssd 3

Medium

Confidence: 97% confidence
Finding: The skill explicitly instructs the agent to ask the user for a third-party Suno API key in chat. Soliciting secrets in plain conversation increases the chance of credential exposure through chat logs, transcript retention, accidental sharing, prompt leakage, or downstream tool/debug logging.

Ssd 3

Medium

Confidence: 98% confidence
Finding: This repeated instruction normalizes disclosure of API credentials directly in the session and ties them to provider identification, making sensitive secret handling part of ordinary conversational workflow. Repetition increases the likelihood users will comply without considering the persistence and visibility of chat contents.

Ssd 3

Medium

Confidence: 98% confidence
Finding: Embedding API-key collection into first-time setup institutionalizes unsafe secret-handling practices. In the context of an agent skill, this is more dangerous because setup flows are high-frequency and likely to be followed verbatim, causing broad exposure of reusable third-party credentials.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: SKILL.md:95