22+ specialized agents that plan, code, test, secure, and deploy software — autonomously. Each agent runs in a sandboxed workspace with real compilers and test suites.
From natural language spec to packaged artifact. Every phase verified with real tools.
Analyses requirements, generates structured plan JSON with phase breakdown and acceptance criteria.
Generates production code across 20+ languages. Creates full directory trees with tests, configs, and docs.
Runs pytest, node --check, go test, cargo test — real compilation and execution in isolated workspaces.
3-round self-healing loop. Reads test failures, patches code, re-runs until green.
Multi-dimensional code review: correctness, security, performance, style. Verdict: pass / request changes.
Generates README, API reference, architecture diagrams, inline comments, and deployment guides.
Tarballs the complete project: source, tests, configs, docker files. Ready to deploy on first boot.
33 security skills across 14 Red Team, 12 Blue Team, and 7 Purple Team specialisations.
Coordinates multi-agent campaigns. Assigns sub-tasks, merges results, tracks progress across the fleet.
Network scanning, port discovery, service fingerprinting, CVE correlation, passive OSINT gathering.
Metasploit API integration, privilege escalation paths, lateral movement, persistence mechanisms.
Credential dumping, screenshot capture, keylogging, data exfiltration, session persistence.
ATT&CK framework mapping, IoC hunting, persistence detection, behavioral anomaly analysis.
Memory dump analysis, timeline reconstruction, artifact extraction, disk image parsing.
CIS benchmark auditing, vulnerability remediation, config hardening, security patch management.
Honeypot deployment, canary token management, trap triggering, attacker distraction.
Domain-specific agents for common development and operations tasks.
Fullstack application generation — CRMs, chatbots, sales pipelines, dashboards from natural language specs.
Automated containment, evidence preservation, runbook execution, post-incident report generation.
Auto-generates and applies safe fixes for critical/high vulnerabilities. Validates before applying.
Attack simulation, detection validation, gap analysis. Generates report comparing expected vs actual detection.
Pre-deployment impact assessment. Analyzes change scope, dependency risks, rollback complexity.
Pre-merge checks: lint, typecheck, test coverage, security scan. Blocks low-quality commits.
Run a real project from a single command line.
python freeai.py auto-start "Build a FastAPI notes service with auth and tests" --watch