agentcage is an open-source CLI tool that provides a defense-in-depth proxy sandbox for AI agents. It uses containerization and MITM proxy techniques to enhance security and isolation, making it suitable for developers and researchers working with autonomous AI agents.
agentcage is an Other AI project. It focuses on securing and isolating AI agents by running them in a proxy sandbox with defense-in-depth measures. It is built as an open-source project for AI developers, security engineers, researchers. agentcage is open source under the MIT license. agentcage is available on the web and the command line.
Behind agentcage is agentcage, and it first shipped in 2026. Development happens publicly on GitHub with 14 stars and 258 commits in the last 90 days. Key capabilities include proxy sandbox, defense-in-depth, and containerization.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do