Phase 01
Expand
Broader agent capability surface
Increase what the agent can do by defining support for the most important, frequently used red-team tools and turning them into reliable task actions.
Roadmap
The v1 release plan for DrowAI
The public MVP proves the core operator-agent workflow. Version 1 is the next target: broader tool and LLM provider support, interactive sessions, faster execution, multi-agent work, continuous assessment, and polished reports.
Version 1 roadmap
The path from public MVP to v1 release
These milestones define what DrowAI needs before it can be treated as a real v1 product: broader agent capabilities and LLM provider support, interactive sessions, faster execution, multi-agent work, continuous assessment, and polished reporting.
Phase 02
Next
Broader LLM provider support
Extend DrowAI's provider layer across leading hosted model providers and local runtimes such as Ollama, so teams can choose models that fit their deployment, privacy, performance, and operational requirements.
Phase 03
Next
Stabilize the end-to-end engagement flow
Polish the current path from task creation through agent work, evidence review, knowledge preparation, and report creation so the full workflow feels dependable and coherent.
Phase 04
Next
Execution speed and tooling optimization
Optimize command orchestration, result handling, and tool execution paths so common assessment loops complete faster and with less operator friction.
Phase 05
Expand
Multi-agent task execution
Support specialized agents that can work on different task types in parallel, such as reconnaissance, exploitation support, and credential attack validation.
Phase 06
Later
Automated continuous assessment
Move from one-off task execution toward recurring assessment flows that can re-check known surfaces and keep engagement knowledge current.
Phase 07
Later
Report polish and export formats
Improve report quality, add stronger review flows, and introduce additional export formats such as PDF-ready engagement reports.