agentic-detector

At this stage, it is better treated as an early open-source detection component worth watching, not a broadly validated endpoint control product. The adoption view should be conservative: the available evidence is essentially the official GitHub repository and its summary. That confirms the project exists and has a clear scope, and a small number of stars/forks shows some attention; however, that is heat proof, not proof of effectiveness in large environments, low false positives, or operational maturity.

Based on the repository description, its practical value is exposing one agentic_software table in osquery so teams can query for MCP servers, AI agent CLIs, AI apps, IDE plugins, and live AI/MCP network sockets in a unified way. It is not a chat assistant, not a code-generation tool, and not an EDR or gateway that directly blocks AI usage. A more accurate analogy is an "osquery detection extension for emerging AI software assets" for teams already running Fleet/osquery.

On cost and setup, the evidence only clearly supports that it is an open-source GitHub project designed for detection-only deployment with Fleet/osquery.