Tool count 131 -> 154; 474 -> 1242 collected tests.
The release exists for two measured reasons. A 149-tool catalog costs a client
~40,300 tokens before it has asked for anything, and a drafter searching the way
drafters think -- FILLET, BPOLY, QSELECT, WBLOCK -- scored df=0 against the stock
index, because those words were simply not in the data.
Discovery and batching
- DISCOVERY_MODE=search replaces the catalog with search_tools + call_tool,
ranked over an AutoCAD command and synonym corpus (discovery/aliases.py) on
top of fastmcp's BM25 index. Measured: 40,305 -> 356 tokens advertised.
- cad_batch runs a step list in one round trip; fields= projects result-heavy
responses. The `core` profile is gone -- discovery replaced it.
Honesty debt (Tier 0)
- OCS/ECS: `extrusion` appeared nowhere in this codebase, so mirrored and
rotated geometry reported coordinates that were quietly wrong. backends/ocs.py
translates at the backend boundary; every tool coordinate is WCS on both
engines, and a tilted plane refuses rather than answering with a number that
looks fine.
- drawing_save_as(fmt="dwg") wrote DXF bytes to a .dwg path. drawing_audit never
repaired. analysis_measure_area shoelaced points the caller typed in.
- ezdxf calls gained a per-call timeout, so one hung call can no longer wedge a
server whose document lock is a single asyncio.Lock.
Architecture
- The 1854-line backend ABC is 19 contract modules; @capability(key, reason=)
makes optional methods declarative and a gate fails on undeclared holes.
- 8 layout/viewport tools, CHSPACE, CAD_PROGID escape hatch.
- Wave A: selection filters, boundary tracing, hatch depth, annotation objects,
measurement -- 13 tools of a planned 39. The REGION and 2D-boolean family was
cut on evidence: add_region() yields zero ACIS bytes and the greiner_hormann
substitute loses 28.2% of the area on a square with one semicircular edge.
Verified against a live AutoCAD 2026, which found four COM defects no amount of
reading would have caught -- AcadPViewport has no ViewCenter, and the line that
set it had been swallowed by a bare except since v1.4.
Benchmark v3 (15 tasks) and what it caught
- v2 scored this server 10/10, which carried no information: every task in it
exercised something the server was built around. Five tasks were added because
they can fail, and three did while being written -- a hatch could not be
measured though the refusal listed HATCH as measurable, the boundary tools and
the measure tool disagreed about the same shape (139.2177 vs 139.2699), and a
circle stored as two bulged vertices measured 0.0.
- Competitor reports are NOT re-scored. They are pinned v1.4 runs against v2 and
were never asked these five questions; the chart shows those rows `not run`.
- Correctness A/B: 21/24 -> 24/24, zero regressions.
Performance, stated plainly
The published v1.4 perf report was recorded on CPython 3.14 and the new one on
3.11.15, so the apparent 2-6x speedup is the interpreter, not this release.
Re-measured on one interpreter, v1.5.0 is 28% slower creating entities and 22%
slower on the 10k roundtrip. Fully attributed: EZDXF_CALL_TIMEOUT=0 restores
v1.4.0's exact numbers. That is the deadlock fix being paid for, the knob is
documented, and winning it back is on the 1.6 roadmap.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
- Delete .github/workflows/ci.yml and not.md.
- Drop the now-inaccurate "GitHub Actions CI" claims from README/CHANGELOG so
the v1.1.0 docs match reality (the ezdxf-in-your-CI capability notes remain).
Three production-blocking defects + two legal/infra gaps from the recon pass.
R1 (CRITICAL): the 8 premium meta-tools (drawing_plan, drawing_critique,
point_from_snap, construction_xline, construction_clear, drawing_apply_iso_layers,
dimension_auto, entity_select_smart) raised NotImplementedError on the COM
backend — i.e. the entire "non-negotiable" premium workflow crashed on the
DEFAULT Windows backend. Lifted all eight to AutoCADBackend as concrete,
backend-agnostic implementations built on the shared abstract primitives +
the identical EntityInfo.properties contract, so both backends inherit them.
The only backend-specific piece left is the new `_create_xline` primitive
(XLINE has no cross-backend entity_create_* equivalent). Removes ~285 lines of
ezdxf-only duplication; the existing meta-tool test suite now covers shared code.
R2 (CRITICAL): layer_create/layer_modify/entity_set_properties used
int(lineweight), truncating every millimetre lineweight (0.50, 0.25, 0.18, 0.13)
to 0 — silently wiping the whole ISO 128 discipline, because engineering/layers.py
passes mm-floats while the MCP boundary passes hundredths-ints. Added
normalize_lineweight() to reconcile both conventions to ezdxf/COM hundredths and
wired it into all six call sites.
R3 (HIGH): _check_iso128 was a permanent no-op (everything was 0 from R2, so the
`lw <= 0: continue` guard skipped all layers). Fixed by R2; additionally exclude
CONSTRUCTION-class layers (intentionally sub-ISO 0.05mm scratch weight) to avoid
a new false positive, and corrected the misleading "normalises to mm" comment.
Also: add MIT LICENSE (referenced by CLAUDE.md but missing — legal gap) and
restore .github/workflows/ci.yml (lint kept non-blocking until the repo-wide
ruff cleanup; the test job is the real gate).
Verified: 182 passed / 2 xfailed; functional check confirms lineweights now
store as hundredths (GEOMETRY=50) and the ISO critique correctly flags a 0.40mm
layer while passing a clean ISO drawing.
