SIFT proposes to design and develop the Marshal system, a mixed-initiative tool for maintaining task models over the course of evolving missions. Marshal-enabled planning and scheduling systems will maintain task models so that they are in tighter correspondence with the current operating context. Marshal will provide a simple dialogue based interface to update models and affordances for incomplete or imprecise user inputs. Marshal will actively diagnose user created plans to identify and rank queries so that users can maintain the most relevant task model constraints. Marshal's capabilities will stem from how it represents and reasons with a family of possible task model interpretations. As users add new constraints that may be incomplete, imprecise, or temporary, Marshal creates interpretations of the constraints. With user feedback, Marshal updates the interpretations to more closely match their intended task model. With a family of models, Marshal can compute the expectation that user plans are consistent. Marshal also computes diagnoses explaining which assumptions about the user's intended model lead to inconsistency. From the diagnoses, Marshal populates a queue of potential plan flaws that the user can address at their convenience through natural and informative dialogues. As users interact with Marshal, it automatically maintains task models so that they can better serve mission planners as missions evolve.