ZeroLeaks offers continuous red teaming for AI agents, endpoints, and MCP tools. It automatically tests for prompt injection, prompt extraction, tool abuse, and other security issues, then provides detailed remediation reports. Teams can run on-demand scans or integrate into CI/CD pipelines. The platform is grounded in a large open-source database of real-world vulnerabilities and is backed by the maintainers of a major prompt-leak research repository.
ZeroLeaks sits in PulseGate's AI & ML category. AI agents and systems being vulnerable to prompt injection, data leaks, and unsafe tool use without systematic testing. It is built as a B2B product for AI developers and security teams. ZeroLeaks follows a freemium model. ZeroLeaks is available on the web, the command line, and API.
It is developed by ZeroLeaks, and it first shipped in 2026. The project is developed in the open on GitHub with 720 stars and 16 commits in the last 90 days. Among its 5 catalogued features are automated red teaming, prompt injection testing, and Continuous CI/CD scanning. It exposes integrations via a public API.
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