gcode

A printer-agnostic workflow for generating, inspecting, dry-running, and statically validating plain FDM .gcode from 3D mesh files by orchestrating real slicer command-line tools. It is intended for slicing mesh formats such as .stl, .obj, unsliced .3mf, .ply, .glb, and .gltf into printer-profiled G-code, for discovering which local slicer backends are installed, and for confirming a mesh is slice-ready or that generated G-code is sound before any printer-specific handoff. It never uploads, starts, or packages print jobs.

The process is driven by a helper script, gcode_tool.py, exposing discover, inspect, slice, and validate subcommands. Every slice requires an explicit wrapper profile JSON that points to an absolute native slicer profile path and carries machine bounds and filament details; the wrapper supplies validation bounds and backend selection while the native profile remains the source of detailed process behavior. Slicing is done as a dry-run first to preview the command, then executed only after the command and profile look appropriate. The preferred backend order is OrcaSlicer, PrusaSlicer, then CuraEngine, and discovery checks both PATH and the usual OrcaSlicer app location. STL, OBJ, and unsliced 3MF are passed directly, while PLY, GLB, and GLTF are converted to temporary STL using trimesh when available.

Validation checks for non-empty content, temperature commands, movement and extrusion moves, XYZ bounds, and unknown-command warnings, and should always run before handing G-code to printer-specific workflows. The skill defines clear boundaries: STEP, DXF, SVG, URDF, and SDF inputs are rejected in v1, and Bambu archive creation and printer contact are out of scope. After producing or modifying a .gcode it hands the explicit file path to a cad-viewer skill when installed, and defers Bambu upload and start workflows to a bambu-labs skill. Reference files cover slicer backends and how to interpret validation output.