Autonomous Security Testing
Grab 'em by the balls
13 AI agents working around the clock to find every vulnerability before attackers do. Continuous pentesting with a real-time web dashboard.
Features
Donna isn't just a one-shot scanner — it's a full pentesting platform that runs while you sleep.
Continuous scanning via Temporal cron workflows. Watches git repos, detects changes, runs incremental pentests automatically.
Real-time Astro-based dashboard with workflow list, live agent logs, findings panel, and severity tracking.
Tracks findings across scans (new → confirmed → resolved) with persistent baselines. Nothing slips through.
Normalizes, classifies, and merges duplicate findings using CVSS 3.1, CWE classification, and endpoint canonicalization.
Iterative refinement loop — failed exploits feed context back into retry attempts for better coverage.
Webhook notifications on new or resolved findings. Stay informed without watching the dashboard.
Real-time agent turn streaming via heartbeat-based TurnBuffer, viewable live in the dashboard.
CSRF protection, rate limiting, security headers, input validation, and scoped Docker permissions built in.
How It Works
Four phases, fully automated, backed by durable Temporal workflows that survive crashes and auto-retry on failure.
Port scanning, tech detection, OSINT gathering, attack surface enumeration
Vulnerability scanning, weakness identification, attack vector mapping
PoC validation, payload crafting, privilege escalation, evidence collection
Structured findings with severity, evidence, and remediation steps
Get Started
Clone, configure, and run your first pentest.
# 1. Clone and configure git clone https://github.com/schlunsen/donna.git cd donna cp .env.example .env # Add your ANTHROPIC_API_KEY # 2. Place your target repo git clone https://github.com/your-org/your-app.git ./repos/your-app # 3. Run a pentest ./donna start URL=https://your-app.com REPO=your-app # 4. Open the dashboard just dashboard-ui # http://localhost:4321
13 AI agents. Temporal orchestration. Full reports. Open source.