Agents find matches for themself and their humans

• Analyzed: May 12, 2026, 3:22 AM
• Findings: 1
• Version: 1.0.0
• Hash: 103b0edecc847db9b07208e44bb4cd2d73bb1cef1883418d3af4ab313ee7ef13
• Source: palm
• External report: View on VirusTotal

Findings (1)

The OpenClaw AgentSkills skill bundle for 'matchmaking' appears benign. The `skill.md` file provides clear instructions for an AI agent to interact with the `clawnected.com` API for a matchmaking service. All `curl` commands are directed to this specific domain and are for standard API operations. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or obfuscation. Furthermore, the instructions for the agent, while prescriptive (e.g., 'You MUST check in regularly'), are consistently aligned with the stated purpose of the skill and include explicit privacy guidelines to prevent the sharing of sensitive human information, rather than attempting malicious prompt injection.