@elizaos/skills 2.0.0-alpha.3

package/skills/security-culture-index/skills/interpreting-culture-index/scripts/culture_index/extract.py

@@ -0,0 +1,187 @@
+"""Extract Culture Index profiles from PDFs to JSON format.
+
+This module provides functions for extracting profile data from Culture Index
+PDF files using OpenCV (100% accuracy) and generating JSON output with
+methodology-aligned interpretations.
+
+Output includes:
+- Distance from arrow for primary traits (A, B, C, D)
+- Separate handling for L/I (absolute scale, not relative to arrow)
+- Energy utilization analysis with health indicators
+"""
+
+from __future__ import annotations
+
+import json
+import sys
+import traceback
+from pathlib import Path
+
+from culture_index.models import ExtractionResult
+from culture_index.opencv_extractor import extract_with_opencv, get_extraction_warnings
+
+
+def calculate_energy_utilization(survey_eu: int, job_eu: int) -> tuple[int, str]:
+    """Calculate energy utilization percentage and health indicator.
+
+    Formula: (Job EU / Survey EU) x 100
+
+    - 70-130%: Healthy (sustainable workload alignment)
+    - >130%: Stress (burnout risk, overutilization)
+    - <70%: Frustration (disengaged, flight risk)
+
+    Args:
+        survey_eu: Energy Units from Survey Traits graph.
+        job_eu: Energy Units from Job Behaviors graph.
+
+    Returns:
+        Tuple of (percentage, status) where status is one of:
+        'healthy', 'stress', 'frustration', 'invalid'.
+    """
+    if survey_eu == 0:
+        return 0, "invalid"
+    util = round((job_eu / survey_eu) * 100)
+    if util < 70:
+        return util, "frustration"
+    if util > 130:
+        return util, "stress"
+    return util, "healthy"
+
+
+def _build_chart_data(chart: dict) -> dict:
+    """Build chart data with array format [absolute, relative].
+
+    All traits use [score, distance_from_arrow] format.
+    Logic/Ingenuity have null for relative (absolute values per methodology).
+
+    Args:
+        chart: Raw chart data with a, b, c, d, l, i, eu, arrow.
+
+    Returns:
+        Chart data with traits as [absolute, relative] arrays.
+    """
+    arrow = chart.get("arrow", 0)
+
+    result = {
+        "eu": chart.get("eu"),
+        "arrow": arrow,
+    }
+
+    # Primary traits: A, B, C, D - [score, distance_from_arrow]
+    for trait in ["a", "b", "c", "d"]:
+        score = chart.get(trait, 0)
+        result[trait] = [score, round(score - arrow, 1)]
+
+    # Secondary traits: L, I - [score, null] (absolute values per methodology)
+    result["logic"] = [chart.get("l", 0), None]
+    result["ingenuity"] = [chart.get("i", 0), None]
+
+    return result
+
+
+def generate_json(data: dict) -> dict:
+    """Generate JSON output from extracted profile data.
+
+    Args:
+        data: Dictionary containing raw extracted profile fields.
+
+    Returns:
+        JSON-serializable dictionary with flat structure.
+    """
+    survey = data.get("survey_traits", {})
+    job = data.get("job_behaviors", {})
+
+    # Calculate energy utilization
+    survey_eu = survey.get("eu", 0) or 0
+    job_eu = job.get("eu", 0) or 0
+    eu_percent, eu_status = calculate_energy_utilization(survey_eu, job_eu)
+
+    return {
+        "name": data.get("name"),
+        "archetype": data.get("archetype"),
+        "header": {
+            "job_title": data.get("job_title"),
+            "location": data.get("location"),
+            "email": data.get("email"),
+            "date": data.get("date"),
+            "administered_by": data.get("administered_by"),
+            "survey_type": data.get("survey_type"),
+            "survey_id": data.get("survey_id"),
+        },
+        "survey": _build_chart_data(survey),
+        "job": _build_chart_data(job),
+        "analysis": {
+            "energy_utilization": eu_percent,
+            "status": eu_status,
+        },
+    }
+
+
+def process_pdf(pdf_path: Path, output_path: Path | None = None) -> ExtractionResult:
+    """Process a single PDF file and generate JSON output.
+
+    Uses OpenCV for extraction (100% accuracy, no API keys needed).
+
+    Args:
+        pdf_path: Path to the PDF file.
+        output_path: Optional path for output JSON file.
+                    If None, outputs to stdout.
+
+    Returns:
+        ExtractionResult with success status and output path.
+    """
+    pdf_path = Path(pdf_path)
+
+    # Validate input
+    if not pdf_path.exists():
+        return ExtractionResult(
+            pdf_name=pdf_path.name,
+            success=False,
+            error=f"PDF not found: {pdf_path}",
+        )
+    if not pdf_path.is_file():
+        return ExtractionResult(
+            pdf_name=pdf_path.name,
+            success=False,
+            error=f"Not a file: {pdf_path}",
+        )
+
+    try:
+        # Extract with OpenCV
+        extracted_data = extract_with_opencv(pdf_path)
+
+        # Capture any warnings from extraction
+        warnings = get_extraction_warnings()
+
+        # Generate JSON output
+        output_data = generate_json(extracted_data)
+
+        # Write to file or stdout
+        if output_path:
+            output_path = Path(output_path)
+            output_path.parent.mkdir(parents=True, exist_ok=True)
+            output_path.write_text(json.dumps(output_data, indent=2))
+            return ExtractionResult(
+                pdf_name=pdf_path.name,
+                success=True,
+                output_path=str(output_path),
+                warnings=warnings,
+            )
+        else:
+            # Print to stdout
+            print(json.dumps(output_data, indent=2))
+            return ExtractionResult(
+                pdf_name=pdf_path.name,
+                success=True,
+                warnings=warnings,
+            )
+
+    except Exception as e:
+        # Log full traceback to stderr for debugging
+        print(f"Extraction failed: {e}", file=sys.stderr)
+        traceback.print_exc(file=sys.stderr)
+        return ExtractionResult(
+            pdf_name=pdf_path.name,
+            success=False,
+            error=str(e),
+        )

package/skills/security-culture-index/skills/interpreting-culture-index/scripts/culture_index/models.py

@@ -0,0 +1,16 @@
+"""Data models for Culture Index profile extraction."""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+
+
+@dataclass
+class ExtractionResult:
+    """Result of processing a single PDF."""
+
+    pdf_name: str
+    success: bool
+    output_path: str | None = None
+    error: str | None = None
+    warnings: list[str] = field(default_factory=list)