#waf #caddy #artificial-intelligence #gateway #security

app gwaf

AI-native WAF for self-hosted apps with deterministic enforcement and reviewable policy workflows

3 releases

Uses new Rust 2024

0.1.5 May 7, 2026
0.1.4 May 7, 2026
0.1.1 May 6, 2026

#1875 in Authentication

AGPL-3.0-only

235KB
5.5K SLoC

Gatewarden

Open-source AI WAF for self-hosted apps.

Gatewarden sits in front of your services, consumes trusted identity headers from your auth layer, applies deterministic enforcement, and keeps AI in a reviewable advisory lane for rule suggestions, event analysis, and operator workflows.

Project repository:

Documentation:

Core capabilities:

  • AI-assisted WAF for self-hosted and internal applications
  • Deterministic protection for admin and login surfaces
  • Trusted-header identity model
  • Caddy-first integration with forward_auth
  • SQLite- or PostgreSQL-backed events, rules, approvals, and settings
  • Web console for events, rules, approvals, settings, status codes, and latency
  • Structured observability from Caddy access logs

Workspace crates:

  • gwaf: main application crate
  • ingress-core: shared request, identity, and event types
  • ingress-gateway: gateway abstractions and adapters
  • ingress-api: console API DTOs
  • ingress-policy: deterministic policy primitives
  • ingress-rate-limit: rate limiting primitives
  • ingress-caddy: Caddy integration helpers

Quick start:

cargo run -p gwaf

The runtime binary name remains:

gatewarden

If you want the full product overview, screenshots, and deployment notes, use the GitHub repository README:

Dependencies

~49–68MB
~1M SLoC