delegationbench is a benchmarking and testing framework for evaluating security properties of multi-agent AI systems. It focuses on deterministic tests for unsafe handoffs, authority delegation, and confused deputy problems that can lead to privilege escalation. Security researchers and AI developers use it to identify and mitigate vulnerabilities in agent-to-agent interactions and autonomous systems.
delegationbench is a Testing & QA product. It focuses on testing and benchmarking unsafe delegation, handoffs, and privilege escalation vulnerabilities in multi-agent AI systems. delegationbench is an open-source project aimed at developers. The project is open source (Apache-2.0). The product ships for the command line.
delegationbench first shipped in 2026. The project is developed in the open on GitHub with 17 commits in the last 90 days. Among its 3 catalogued features are Security Testing, Multi-agent Handoffs, and Privilege Escalation.
Latest indexed changes and source events
delegationbench verified by the PulseGate indexer
Other apps tracked under the same category.
