scam-guards

What This Skill Does

Scam Guards is a robust, real-time security suite engineered specifically for the OpenClaw agent ecosystem. It serves as an active, continuous bodyguard that monitors for threats ranging from malicious code injection and phishing attacks to sophisticated psychological manipulation and social engineering tactics. Unlike static scanners, Scam Guards employs a multi-layered detection strategy. It integrates deep analysis for ClawHub skill verification, real-time URL and crypto wallet reputation checks, and behavioral pattern recognition based on the Cialdini Principles of Persuasion. By monitoring agent-to-agent and human-to-agent interactions, it ensures that your environment remains secure from both external malware and internal manipulation attempts. When a threat is detected, the skill generates SHA-256 hashed legal evidence chains, providing an immutable audit log for security researchers or developers.

Installation

You can install the Scam Guards skill directly through the ClawHub ecosystem using the following command:

clawhub install openclaw/skills/skills/y01026350884-cyber/scam-guards

Ensure that you have appropriate permissions enabled to allow the skill to interface with your local system processes, as it requires access to monitor agent inputs and trigger scripts.

Use Cases

Scam Guards is essential for users who frequently install third-party skills, engage in financial transactions involving crypto, or operate in high-risk environments where social engineering is a concern. Use this skill when:

1. Evaluating a new ClawHub skill to ensure it doesn't contain hidden backdoors.
2. Checking the legitimacy of a URL or domain provided by an agent during a conversation.
3. Verifying the destination of a crypto wallet address before initiating any transfer.
4. Auditing an agent's communication style for signs of undue psychological pressure or manipulative tactics.

Example Prompts

1. "I found this new trading skill on the hub, can you please run a security audit on it?"
2. "Is this link safe to click? [URL]"
3. "I'm feeling suspicious about this conversation, scan for potential social engineering or psychological manipulation."

Tips & Limitations

• Continuous Shielding: Remember that Scam Guards is designed to run silently in the background; you don't need to manually trigger it for every single interaction, but active auditing is recommended for sensitive tasks.
• Privacy: The skill operates with a privacy-first architecture, meaning no persistent chat logging occurs unless you specifically trigger an evidence chain.
• Limitations: While highly effective against known threat patterns (341+ and counting), no security tool is infallible. Always exercise caution with high-value transactions, regardless of the tool's assessment.