MCP Optimizer

from __future__ import annotations from crewai_tools import BaseTool from ..schemas import LPModel, SolveOptions from ..solvers.mip.branch_and_cut import solve_mip class SolveMIPTool(BaseTool): name = "solve_mixed_integer_program" description = ( "Solve a mixed-integer linear program using a depth-first branch and bound search." ) def _run(self, model: dict, options: dict | None = None, use_or_tools: bool = False) -> dict: mip_model = LPModel.model_validate(model) opts = SolveOptions.model_validate(options or {}) solution = solve_mip(mip_model, opts, use_or_tools=use_or_tools) return solution.model_dump()