GatekeeperAI v1.0.0 — MVP Release

GatekeeperAI v1.0.0

The first production-ready release of GatekeeperAI — an on-premises AI application governance platform for enterprise security teams.

What's included

Core platform

• Full scan pipeline: secrets detection, dependency audit, PII scanning, license compliance, and AI risk analysis powered by Claude
• Risk tier engine (Green / Yellow / Red) with SLA-based approval routing
• Approval workflow with reviewer queue, decision recording, and email notifications
• Container-based app deployment with isolated runtime and secrets injection
• JWT auth with refresh token rotation, rate limiting, and security headers

Admin & operations

• First-run setup wizard — no config files to edit before first login
• Admin panel: user management, role assignment, audit log viewer
• Secrets manager: per-app encrypted secret storage surfaced in the IC dashboard
• SLA monitoring via Celery Beat with automated escalation

Developer experience

• Built-in SSH git server (port 2222) — push to main, scan starts automatically
• Clone URLs and git remote commands shown directly in the dashboard
• Deployment detail page with live log streaming and admin controls

Observability

• Structured audit log forwarding to: webhook/HTTP, syslog (RFC 5424), Grafana Loki, AWS CloudWatch
• All sinks are opt-in via .env — zero config required to run

Packaging

• Single docker compose up --build boots all 7 services (postgres, redis, api, worker, beat, frontend, git)
• Smoke tested: all services healthy, migrations run automatically on first boot
• .env.example documents every setting

Installation

See INSTALL.md for step-by-step instructions including cloud hosting (AWS, Azure, GCP) and custom domain setup.

Default ports

Service	Port
Frontend	3000
API	8000
SSH git server	2222
PostgreSQL	5433
Redis	6379