Add docs, scripts, skills, commands, and taskmaster config to repo

Includes:
- docs/: architecture, block-schema, migration-plan, product-specification
- scripts/: bidi_table, decompose-decisions, extract-claims, seed-knowledge, etc.
- skill-legal-decision/: SKILL.md + references + block-schema
- skill-legal-assistant/: SKILL.md
- skill-legal-docx/: SKILL.md + references
- .claude/commands/: bidi-table skill
- .taskmaster/: task config + PRDs
- .gitignore: exclude legacy/, kiryat-yearim/, node_modules/, memory/

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

scripts/bidi_table.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""BiDi-safe box-drawing table renderer for mixed Hebrew/English terminal output.
+
+Uses LRM (Left-to-Right Mark, U+200E) before box-drawing characters to prevent
+the BiDi algorithm from breaking table alignment when Hebrew text is present.
+
+Usage as module:
+    from scripts.bidi_table import bidi_table
+    print(bidi_table(['Col1', 'Col2'], [['val1', 'ערך2']]))
+
+Usage from CLI:
+    python3 scripts/bidi_table.py
+"""
+
+from __future__ import annotations
+
+LRM = "\u200E"  # Left-to-Right Mark — invisible, prevents BiDi reordering
+
+
+def bidi_table(headers: list[str], rows: list[list[str]]) -> str:
+    """Render a box-drawing table safe for mixed RTL/LTR terminal display."""
+    ncols = len(headers)
+
+    # Calculate column widths
+    col_widths = [len(h) for h in headers]
+    for row in rows:
+        for i, cell in enumerate(row[:ncols]):
+            col_widths[i] = max(col_widths[i], len(cell))
+
+    def hline(left: str, mid: str, right: str) -> str:
+        return left + mid.join("─" * (w + 2) for w in col_widths) + right
+
+    def dataline(cells: list[str]) -> str:
+        parts = []
+        for i in range(ncols):
+            cell = cells[i] if i < len(cells) else ""
+            padded = cell + " " * max(0, col_widths[i] - len(cell))
+            parts.append(" " + padded + " ")
+        return LRM + "│" + (LRM + "│").join(parts) + LRM + "│"
+
+    lines = [hline("┌", "┬", "┐")]
+    lines.append(dataline(headers))
+    lines.append(hline("├", "┼", "┤"))
+    for row in rows:
+        lines.append(dataline(row))
+    lines.append(hline("└", "┴", "┘"))
+    return "\n".join(lines)
+
+
+if __name__ == "__main__":
+    table = bidi_table(
+        ["File", "Description", "Model", "Step"],
+        [
+            ["claims_extractor.py", "חילוץ טענות מכתבי טענות", "Sonnet", "שלב 3 — הבא בתור"],
+            ["brainstorm.py", "סיעור מוחות — כיווני נימוק", "Sonnet", "שלב 4"],
+            ["block_writer.py", "כתיבת בלוקים של החלטה", "Sonnet/Opus", "שלב 5"],
+            ["qa_validator.py", "בדיקת איכות QA", "Sonnet", "שלב 6"],
+            ["style_analyzer.py", "ניתוח סגנון דפנה", "Opus", "חד-פעמי"],
+            ["learning_loop.py", "למידה מהחלטה סופית", "Sonnet", "סוף תהליך"],
+        ],
+    )
+    print(table)