Open Apache-2.0 corpus, runner, scoring, and result-verification contracts for measuring AI-agent egress controls. A PipeLab open project.
-
Updated
Nov 14, 2026 - Go
Open Apache-2.0 corpus, runner, scoring, and result-verification contracts for measuring AI-agent egress controls. A PipeLab open project.
Provide an open test corpus to assess and improve AI agent egress security through validated cases and automated workflows.
GitHub Action for running agent-driven CI jobs through Pipelock with verifiable egress evidence.
High-performance Squid caching and egress forward proxy configuration for OpenShift clusters
Deny-by-default Python HTTP(S) egress guard with SSRF prevention, DNS rebinding defenses, TLS validation and strict request limits.
To associate your repository with the egress-security topic, visit your repo's landing page and select "manage topics."