feat: strengthen CAD experience distillation and routing

cad-experience-plugin/skills/cad-experience-builder/SKILL.md

@@ -25,14 +25,18 @@ published library and require repeated support before promoting any experience.
    ```
 
 4. Read `work/review/semantic-batch.json` completely. Act as the semantic
-   distiller: compare families, features, relations, prior vocabulary, failure
-   rules, and validation targets; then author
+   distiller: compare families, parameter roles, datum roles, canonical
+   reconstruction stages, feature cardinality classes, relations, prior
+   vocabulary, failure rules, and validation targets; then author
    `work/review/experience-draft.json` using
    `references/llm-review-contract.md`.
 
-   Propose methods and invariants, not raw co-occurrence pairs. Do not infer
-   operation order, manufacturing intent, or failure causes that final B-Rep
-   evidence cannot support.
+   Propose methods, invariants, and family-level `reconstruction_grammar`
+   records, not raw co-occurrence pairs. A reconstruction grammar is a
+   canonical backend-neutral strategy inferred from repeated geometry; it is
+   not the source model's recovered feature history. Do not infer
+   manufacturing intent or failure causes that final B-Rep evidence cannot
+   support.
 
 5. Publish only after the model draft exists:
 
@@ -70,10 +74,15 @@ required.
 
 ## STEP extraction boundary
 
-- The extractor reads analytic B-Rep surfaces, bounding boxes, topology counts,
-  axes, radii, and locations into a private evidence record.
-- It derives only semantic candidates such as rotational stacks, axis
-  relationships, transitions, and repeated-hole motifs.
+- The V2 extractor reads analytic B-Rep surfaces, bounding boxes, topology
+  counts, axes, radii, and locations into a private evidence record.
+- It derives scale-independent body classes, part families, rotational and
+  prismatic motifs, passage and pattern relationships, complexity roles, and
+  normalized ratio observations before sanitization.
+- It also derives private exact cardinality evidence plus sanitized symbolic
+  parameter roles, datum roles, feature-count classes, and canonical
+  reconstruction stages. These are the evidence used to learn how to rebuild a
+  family without publishing any teacher dimensions.
 - Exact source paths, hashes, dimensions, coordinates, and face references are
   permitted in private case JSON and forbidden in the published library.
 - Keep input/output as a human-visible staging area and induction corpus. Never
@@ -86,7 +95,8 @@ required.
 - One case is evidence, not experience.
 - Publish generalized, dimension-free observations below threshold as visible
   `candidate_experiences`, but never expose them to generation consumers.
-- Require `min_support` independent cases and `min_confidence` before promotion.
+- Require `min_support` independent cases and conditional `min_confidence`
+  within the proposal's semantic context before promotion.
 - Promote feature compositions, semantic constraints, dimensionless
   distributions, validation checks, and failure-repair methods.
 - Reject absolute dimensions, coordinates, face references, source paths,
@@ -105,6 +115,8 @@ The library is backend-neutral. A consumer receives:
 - `dimensionless_distribution`: multi-case ratio ranges;
 - `validation_rule`: checks independent of one part's coordinates;
 - `failure_repair`: reusable detection and repair methods.
+- `reconstruction_grammar`: symbolic parameters, datums, canonical feature
+  stages, optional/required feature roles, and validation roles for a family.
 
 build123d, SimpleCADAPI, or another backend translates the same queried
 experience into its own operations. Experience must never contain executable