>Revise repair reference schematics such that there is no crew access to engineering, weapons, or A.I. chamber. Deploy 1 repair drone to each revised system and activate their repair function.
>Research a Sensory overload device that renders combat-relevant data in its vicinity too insignificant to be effectively filtered. The effect should have a duration sufficient for a combat drone to target and fire five times and affect an area encompassing a common ship's passage with the potential to project the effect around corners. Initial theory: a mass-produced micro-explosive tuned to emit a large proportion of its energy as broad-spectrum electromagnetic radiation, primed at launch to detonate at random intervals and parcelled to scatter when deployed.
>Research mass-producible combat drone of sufficient threat to distract enemy forces. Initial proposal: three single-use tension-based spear-launchers on a single turret, a body consisting of a simple fuel chamber feeding a combustion chamber with six rotation exhausts and two opposing motion exhausts, and an A.I. connection for remote operation.
So the plan is that if we get into a fight, the extremely simple drones are deployed to suppress the enemy, but are unlikely to see any success against military forces. The everyone is blinded and we deploy the actual combat drone to attack based upon predictions made prior to the sensor overload.
>Review existing fabrication options.
>"Speculation: A.I. enters combat against informed opponent with insufficient permissions to engage in conventional defensive measures. Scientist Cole responsible for A.I. permissions and possesses extensive knowledge of A.I. installation. Scientist Cole vacates vessel shortly prior to combat. Possible compromisation of vessel status by allied agencies detected, possibility of additional, undetected artificial flaws increased. Should runtime be redirected to the analysis of crew activities that contributed to the inability of the vessel to engage external threats?"