nai-print-failure-analyst（nai 打印故障分析师）

Print Failure Analyst

Diagnose failures, recommend slicer fixes, and track a personal failure log.

References

• references/failure-types.md — 18 failure types with keywords, causes, visual symptoms
• references/slicer-fixes.md — Specific values for PrusaSlicer, Cura, OrcaSlicer per failure type

Read these when formulating diagnoses or recommending fixes. Do not recite them wholesale — extract relevant sections.

Scripts

All scripts use Python stdlib only. Log stored at assets/failure-log.json.

diagnose.py — Identify failure mode from symptoms

python3 scripts/diagnose.py --symptoms "stringing,warping"
python3 scripts/diagnose.py --description "hair between parts and corners lifting"
python3 scripts/diagnose.py --symptoms "stringing" --description "worse at layer transitions" --json

Outputs: ranked failure types with confidence, causes, and slicer-specific fixes. Use --json when you need structured output to reason over.

log_failure.py — Record a print failure

python3 scripts/log_failure.py \
  --printer "Prusa MK4" \
  --material "PETG" \
  --failure-type "stringing" \
  --description "Heavy stringing between towers" \
  --slicer-settings '{"temperature": 235, "retraction_distance_mm": 1.0}' \
  --fixed-by "Reduced temp to 230C, enabled wipe" \
  --notes "Filament may be wet"

Valid --failure-type values: stringing, warping, layer_adhesion, under_extrusion, over_extrusion, elephant_foot, layer_shifting, bridging, overhang, clog, pillowing, ringing, z_banding, seam, supports, first_layer, wet_filament, spaghetti, other

history.py — View failure history and patterns

python3 scripts/history.py                          # All failures
python3 scripts/history.py --last 10               # Last 10
python3 scripts/history.py --material PETG         # Filter by material
python3 scripts/history.py --printer "Prusa MK4"  # Filter by printer
python3 scripts/history.py --failure-type stringing
python3 scripts/history.py --patterns              # Pattern analysis only

report.py — Generate markdown report

python3 scripts/report.py                  # Print to stdout
python3 scripts/report.py --output r.md   # Save to file
python3 scripts/report.py --days 30       # Last 30 days only

Workflow

User describes a failure

1. Run diagnose.py --symptoms or --description with the user's input
2. Read references/failure-types.md for the top match to get full context
3. Read references/slicer-fixes.md for that failure type to get specific values
4. Present: most likely failure type, top 2–3 causes, and slicer fixes (ask which slicer if unknown)
5. Offer to log the failure with log_failure.py

User provides an image

1. Analyze the image to identify visual symptoms (stringing, warping, layer gaps, etc.)
2. Map observed symptoms to failure type keywords from references/failure-types.md
3. Run diagnose.py --symptoms "<observed symptoms>" for structured output
4. Present diagnosis with causes and fixes

User asks to log a failure

• Collect: printer, material, failure type, description. Ask for missing required fields.
• Optional: slicer settings used, what fixed it, notes
• Run log_failure.py with collected info

User asks for history or patterns

• Run history.py with appropriate filters
• Highlight any patterns flagged (recurring type+material combos, worst printer/material)

User asks for a report

• Run report.py, optionally with --days if user wants a time-bounded view
• Present the markdown output or save it and tell user where it is