agent-bouncer is an open-source policy enforcement package for controlling AI agent spending. It evaluates transactions against declarative policies, blocks violations, and creates tamper-evident signed audit logs for developers building agent-based payment systems.
agent-bouncer sits in PulseGate's Security & compliance platforms category. It focuses on preventing AI agents from making unauthorized or noncompliant transactions without an auditable record. It is built as an open-source project for developers building AI agent payment and authorization systems. The project is open source (MIT). agent-bouncer is available on the command line, and it can be self-hosted.
Behind agent-bouncer is nmaltese13, and it first shipped in 2026. Development happens publicly on GitHub with 23 commits in the last 90 days. Among its 6 catalogued features are declarative policies, transaction blocking, and spending controls.
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