Skip to content

Techstack

Operational source:

C:\Users\adeel\OneDrive\100_Knowledge\203_TextCAD\01_Product_Project_Management\00_Project_Management_n_skills\01_tracks\techstack

This page mirrors the detailed operational tracking for this track so the deployed wiki shows the same score tables as the source TRACKING.md.

DraftLint naming update

DraftLint now names a system family, not one backend. Use the DraftLint System Family map to distinguish the upstream code, Railway Validator, IONOS Detector Gateway, RapidDraft Drawing Analysis product, model-training operations, and historical AMD/Roboflow lab.

Last updated

2026-04-15

Scoring model

This track uses milestone and feature scorecards.

Score guide:

  • 100 means complete, validated, and stable for current scope
  • 90-99 means strong and working, but still carrying meaningful gaps
  • 75-89 means materially working, but still missing important completeness
  • 50-74 means real foundation exists, major parts are still absent
  • below 50 means early, partial, or mostly planned

For this code-heavy platform track, 100 requires implementation plus clear runtime or integration evidence.

Evidence sources reviewed

  • C:\Users\adeel\OneDrive\100_Knowledge\203_TextCAD\01_Product_Project_Management\00_Project_Management_n_skills\01_tracks\techstack\MASTER_PLAN.md
  • D:\02_Code\33_DraftLint_Backend\README.md
  • D:\02_Code\33_DraftLint_Backend\TODO.md
  • D:\02_Code\34_RapidDraft_CADWorkbench_Service\README.md
  • D:\02_Code\47_NXconnection\README.md
  • D:\02_Code\47_NXconnection\NATIVE_CPP_HOST_ARCHITECTURE.md
  • D:\02_Code\47_NXconnection\native\NxLiveHostCpp.cpp
  • D:\02_Code\47_NXconnection\scripts\build-nx-live-cpp.ps1
  • D:\02_Code\39_radical_tauri_selfdev\plans\MASTER_PLAN.md
  • C:\Users\adeel\OneDrive\100_Knowledge\203_TextCAD\01_Product_Project_Management\TextCAD_Wiki\docs\06_Infrastructure_Research\_INDEX.md

Current overall score

Techstack overall score: 72 / 100

Why the score is not higher: There is now more than generic platform plumbing here: the managed NX live host can drive the visible NX session and the native NX Open C++ host has been brought up as a real kernel-stage prototype. The score is not higher because the native host still lacks full drawing-loop parity and section-view automation remains the hardest NX-specific gap.

Milestone scorecard

Milestone Score / 100 Current state What it helps achieve Main remaining gap
1. Backend and service map 60 Partial to working Clarifies what each runtime component is responsible for. A single end-to-end platform map is still missing.
2. Typed-tool boundary 79 Working Keeps Studio orchestration and agent actions explicit. Tool/interface contracts are stronger, but still not fully centralized.
3. CAD desktop service integration 84 Working Gives the shell live desktop CAD capability through a controllable runtime surface. The native host has not yet reached drawing-loop parity.
4. Runtime environment stability 70 Foundation exists Keeps local execution and session targeting predictable. Multiple host modes and NX-specific environment assumptions still exist.
5. Shared model/data path governance 55 Early to partial Makes shared stores and provenance manageable. Policies still live more in code and READMEs than in one operating contract.
6. Diagnostics and hardening 61 Working but incomplete Makes failures more understandable and easier to recover from. NX section-view associativity is still a brittle edge case.
7. Native NX host migration 58 Foundation exists Creates the long-term high-access control path for the visible NX session. Only the kernel commands are native today; review-loop commands are not yet ported.

Feature scorecard

Feature track Score / 100 What it helps achieve Current note
Backend runtime 74 Gives the product a functional service backbone. Stronger than before, though still not fully consolidated.
CAD desktop service 84 Bridges desktop CAD capability into the product. The managed NX live host now proves real direct session control.
NX live control 68 Lets the agent drive the exact visible NX session instead of detached helpers. Managed path works; native path is still kernel-stage.
Typed tools 75 Makes AI actions explicit and bounded across transports. Shared JSON contracts now span more than one host model.
Service contracts 64 Prevents ambiguity between runtime components. Much clearer, but still not documented as one complete platform map.
Environment portability 58 Makes local setup less brittle. Session registration and build scripts help, but NX-specific assumptions remain.
Diagnostics 60 Makes failures understandable and repairable. Better than before, though section failures remain hard.
Platform documentation 67 Helps future maintainers and agents reason about the runtime. The NX live-host story is now materially documented, not just implicit in code.

Current headline assessment

Techstack now includes a real live-control path into desktop CAD, which is a meaningful step up from generic infrastructure plumbing. The managed NX host can already drive the visible NX session and the review-loop commands, while the native NX Open C++ host has crossed from plan to compiled kernel prototype. The current score reflects that this is materially real now, but the native migration is still incomplete and the hardest drafting edge cases are not yet solved.

Highest-value next actions

  1. Port the review drawing loop from the managed host into the native NX Open C++ host.
  2. Solve section-view associativity natively or through Open C / UFUNC fallback.
  3. Publish one platform map that spans DraftLint, CAD Workbench Service, Studio, and NX live control.

Sources

  • C:\Users\adeel\OneDrive\100_Knowledge\203_TextCAD\01_Product_Project_Management\00_Project_Management_n_skills\01_tracks\techstack\MASTER_PLAN.md
  • C:\Users\adeel\OneDrive\100_Knowledge\203_TextCAD\01_Product_Project_Management\00_Project_Management_n_skills\01_tracks\techstack\TRACKING.md