Feasibility as a core task property
A useful agentic task must be feasible. At least one trajectory in the current environment must satisfy the user prompt while respecting the system specification. This excludes tasks that require unavailable tools, inaccessible knowledge, impossible state transitions, or policy-prohibited actions.
Evidencia a favor
AutoSynthData: Generating Training Data for Enterprise Agents
Extracto original
Feasibility. There should exist at least one trajectory in the current environment that satisfies the user prompt while respecting the system specification. This rules out tasks that depend on unavailable tools, inaccessible knowledge, impossible state transitions, or actions prohibited by policy.