augura 0.1.3

BREP-exact 3D-printability analysis and print-advice for build123d / FDM

augura

Will it print? Ask before you print.

augura analyses a 3D part and tells you how it will print — before you slice it. It reads the exact boundary representation (OpenCASCADE, via build123d) rather than a tessellated mesh, so overhang angles, wall thicknesses, and the contact footprint are computed analytically — not approximated. Point it at a STEP file and get a deterministic, explainable printability report.

augura — Spanish/Italian, "it foretells".

Capabilities

Check	What it flags
overhangs	downward faces shallower than the support angle (bed contact excluded)
orientation search	ranks candidate poses by unsupported overhang area
manifold / watertight	a part that isn't a closed solid
tip-over	centre of mass projecting outside the bed-contact footprint
bed-fit	a part too big for a declared build volume
brim / raft	small-footprint / high-aspect parts likely to lift
minimum feature	the smallest vertical step → your maximum layer height
wall thickness	walls thinner than min_perimeters × nozzle

Inputs: STEP / build123d solids (exact) · STL meshes (degraded, approximate).

Install

pip install augura      # or: uv pip install augura

STL support is built in — no extra needed.

CLI

# Full printability report (human-readable)
augura analyze bracket.step

# Markdown (drop into a PR or wiki) or JSON (machine-readable)
augura analyze bracket.step --format md
augura analyze bracket.step --format json

# Analyse at the best print orientation instead of as-imported (works with
# any --format; the rotation used is reported in the output)
augura analyze bracket.step --best-orientation

# ... or at a pose you chose yourself, e.g. from `augura orientations`
augura analyze bracket.step --orient 90 0 0

# estampo.toml fragment (feed into an estampo [[parts]] entry);
# with --best-orientation it emits orient = [X, Y, Z] for the top-ranked pose
augura analyze bracket.step --format estampo --best-orientation

# Tune to your printer / job
augura analyze bracket.step --nozzle 0.6 --min-perimeters 3
augura analyze bracket.step --build-volume 256 256 256 --support-angle 50

# Gate a build: exit non-zero if any ERROR-severity issue (great for CI / pre-slice)
augura analyze bracket.step --exit-code

# Rank print orientations by how much support each needs (STEP only);
# --build-volume ranks poses that fit first and marks those that don't
augura orientations bracket.step --format json --build-volume 256 256 256

augura analyze accepts .step / .stp (exact BREP path) or .stl (mesh path).

Example — text:

augura — bracket.step
  [WARNING] overhang: Downward face at 0.0 deg from horizontal needs support
  [WARNING] brim: Recommend a brim: small bed-contact footprint (28 mm2)
  2 finding(s): 0 error, 2 warning

Example — JSON (--format json):

{
  "source": "bracket.step",
  "findings": [
    {"kind": "overhang", "severity": "warning",
     "message": "Downward face at 0.0 deg from horizontal needs support",
     "area": 250.0, "location": [25.0, 0.0, 50.0]}
  ]
}

Flag	Meaning
--format text|md|json|estampo	output format (default text); estampo (analyze only) emits an estampo.toml fragment
--best-orientation	rotate to the top-ranked print orientation before analysing (poses that fit --build-volume rank first); the rotation is reported in the output (STEP input only)
--orient X Y Z	analyse at this explicit Euler rotation (degrees) instead — e.g. a pose chosen from augura orientations (STEP input only)
--support-angle	overhang threshold, degrees from horizontal (default 45)
--nozzle / --min-perimeters	wall-thickness limit = min_perimeters × nozzle
--build-volume X Y Z	enable the bed-fit check against this volume (mm)
--exit-code	exit 1 if any ERROR finding (manifold / bed-fit / tip-over)

Python API

For build123d users, augura is a library too — one façade, analyze(), returning a Report:

import augura
from build123d import import_step

report = augura.analyze(import_step("bracket.step"))

for f in report.findings:
    print(f.severity, f.kind, f.message)

report.overhangs        # tuple[Finding, ...] — just the overhang findings
report.bed_fit          # likewise per check: .manifold_issues, .tip_over,
                        # .brim, .min_feature, .thin_walls
report.to_dict()        # JSON-serialisable view

Pass any build123d Shape you already have in memory — no file needed:

from build123d import Box
report = augura.analyze(Box(40, 40, 5), build_volume=(256, 256, 256))

Individual checks and the orientation search are public too:

augura.find_overhangs(shape, support_angle=45.0)
augura.find_thin_walls(shape, nozzle=0.4, min_perimeters=2)
augura.orientation_scores(shape)        # ranked list[OrientationScore], best first
augura.apply_orientation(shape, (180, 0, 0))  # pose a shape as a score was ranked
augura.is_watertight(shape)             # bool

Top-level surface: analyze, analyze_mesh, apply_orientation, Report, Finding, Severity, OrientationScore, find_overhangs, find_bed_fit, find_manifold_issues, find_tip_over, find_brim_risk, find_thin_features, find_thin_walls, min_vertical_feature, min_wall_thickness, is_watertight, orientation_scores, DEFAULT_SUPPORT_ANGLE.

Where it fits

build123d-mcp  →  augura          →  estampo            →  bambox        →  printer
(design)          (will it print?)   (reproducible slice)  (Bambu package)

augura is loosely coupled — it depends on none of its siblings and is usable on its own. Other tools wrap it; it wraps nothing.

Status

Pre-alpha. The public API is still evolving; 0.1.x may introduce breaking changes. See SPEC.md for the mission, detailed aims, and non-aims, and CHANGELOG.md for releases.

License

Apache-2.0.