Bambu Labs

Bambu Labs

Provenance: maintained in earthtojake/text-to-cad. Use the installed local skill files as the runtime source of truth; the repository link is only for provenance and release review.

Use this skill for local-network Bambu Lab print handoffs after a plain .gcode file already exists and has been validated. This skill does not slice models.

Safety Rules

• Default to dry-run plans. Real printer traffic requires --execute.
• Never start a print without --execute --confirm-start-print.
• Pause and cancel controls are live printer requests; default to dry-run plans. Canceling a print requires --execute --confirm-cancel-print.
• Treat an explicit user request to print or start a specific job as live-start authorization; do not pause for a second confirmation solely for physical checks. Still validate the G-code, inspect the dry-run payload, read printer status, prefer upload-only before upload-start, state the physical checks, and stop if validation/status/intent is unsafe or ambiguous.
• Do not ask for the printer serial by default; fetch it from the printer TLS certificate with serial or let send cache it.
• Prefer workspace-root bambu-printers.json over repeating access codes in commands. The file is local config and should be ignored by Git.
• Before a live start, state the physical checks: clear build plate, correct plate/filament/nozzle, safe surroundings, and operator nearby.
• Publishing MQTT is only a start request. Confirm acceptance with printer status/UI and physical observation.

CAD Viewer Handoff

After completing Bambu work that creates or modifies a local .3mf print artifact, you must ALWAYS hand the explicit file path to $cad-viewer when that skill is installed. CAD Viewer does not open .gcode, so a plain G-code artifact takes no handoff. $cad-viewer must start CAD Viewer if it is not already running and return link(s) to the relevant created or updated file(s); if $cad-viewer is unavailable or startup fails, report that instead of silently omitting the handoff.

Workflow

1. Generate and validate plain G-code with $gcode. If no slicer is installed, install OrcaSlicer and retry; do not treat the missing slicer as a blocker. On macOS, prefer brew install --cask orcaslicer.
2. Configure the printer. The user can either give the IP/access code in the thread and let the agent write JSON, or edit bambu-printers.json directly. For a new printer setup or onboarding request, read references/new-printer-onboarding.md first. Walk the user through the model-specific touchscreen steps to find the IP and LAN access code, and make Enable LAN Only plus Enable Developer Mode explicit before running local start workflows.

python scripts/bambu_lan_print.py config set \
  --printer a1-mini \
  --host 192.168.1.34 \
  --access-code 12345678 \
  --model a1-mini \
  --fetch-serial

Manual JSON shape:

{
  "printers": {
    "a1-mini": {
      "host": "192.168.1.34",
      "access_code": "12345678",
      "model": "a1-mini"
    }
  }
}

On A1/A1 Mini, find the IP and LAN access code on the printer touchscreen under network/LAN settings. Enable LAN Only and Developer Mode when offered, then power-cycle before retrying local start commands.

3. Read status before live work:

python scripts/bambu_lan_print.py status \
  --printer a1-mini \
  --push-all \
  --wait-seconds 10

4. Dry-run the exact handoff, inspect the JSON payload, then run upload-only. Only after upload succeeds should you run upload-start. If the user explicitly asked to print or start the job, proceed to upload-start --execute --confirm-start-print after the validation, status, and upload checks pass. If the user only asked to prepare, slice, upload, or review, stop before the start request.