Sma Dashboard Generator

SMA Dashboard

Interactive dashboard for Snowflake Migration Accelerator (SMA) analysis results.

Modules

Requirements

The script expects the Reports/ folder to be in the root of the workload directory with at least one of:

• Reports/Issues.csv (with data rows)
• Reports/ArtifactDependencyInventory.csv (with data rows)

IMPORTANT:

• sma-output/Reports/ is NOT supported. The Reports/ folder must be directly in the root.
• This tool ONLY creates/uses sma_storage.sqlite3. It will NOT modify any other SQLite files.

Note: The dashboard will be generated with only the modules for which data is available:

• EWI Tracker and File Tracker: Require Issues.csv with data rows
• Dependency Tracker: Requires ArtifactDependencyInventory.csv with data rows

If a CSV file exists but only contains the header (no data rows), that module will show "No data available".

Output Format

Every time you begin a step, sub-step, or significant action, prefix the message with a timestamp in the format [YYYY-MM-DD HH:MM:SS]. Obtain the current time by running date '+%Y-%m-%d %H:%M:%S' in bash.

Example:

[2026-03-24 14:05:32] Starting Step 1: Detect SMA output path...
[2026-03-24 14:05:33] Found SMA output at /path/to/output
[2026-03-24 14:05:34] Step 1 complete.

Usage

Step 0: Initialize Git

Ensure the workload directory has a git repository on the sma/migration-process branch. This is idempotent — if the orchestrator already initialized git, this is a no-op.

result = sma_api.git_ensure_ready("<workload_path>")

Step 1: Ask User for Workload Path

Ask user for the path to the SMA output directory:

• Use ask_user_question tool with type "text"
• Question: "Path to SMA output directory (containing Output, Reports, Logs folders)"
• This directory contains:
  • Output/ - Converted .py/.ipynb files
  • Reports/ - Issues.csv or IssuesConnect.csv
  • Logs/ - Conversion logs

Step 2: Verify Requirements

Once you have the workload path, verify the data source exists:

# Check for Reports/ folder in root (NOT sma-output/Reports/)
ls "<workload_path>/Reports/Issues.csv" 2>/dev/null || \
ls "<workload_path>/Reports/ArtifactDependencyInventory.csv" 2>/dev/null

If none exist, inform the user that no valid data source was found. Remind them that Reports/ must be in the root directory.

Step 3: Run the SMA Dashboard Manager

cd "<workload_path>" && python3 "<skill_directory>/scripts/sma_manager.py" --template-dir "<skill_directory>/templates"

Where:

• <workload_path> is the path provided by the user
• <skill_directory> is the directory where this SKILL.md file is located

IMPORTANT: Do NOT pass a csv_path argument. Let the script find data sources automatically.

The script will:

1. Auto-detect the data source (SQLite or CSV)
2. Parse and extract unique EWI codes
3. Process ArtifactDependencyInventory.csv if available
4. Calculate dependency islands (connected components)
5. Create sma-dashboard/ folder with:
   • manifest.json - Dashboard configuration and module status
   • index.html - Main dashboard with side panel navigation
   • server/sma_server.py - Standalone server script
   • assets/styles.css - Stylesheet
   • ewi-tracker/ - EWI tracker, File tracker, and Dependency tracker modules
6. Print the summary
7. Start the dashboard server and open in browser

The server will keep running until the user presses Ctrl+C. While running:

• Changes to status/notes are saved automatically to SQLite
• The dashboard stays in sync with the database

Step 4: Report results

The script outputs a complete summary including:

• List of generated files
• Total EWIs and occurrences
• Breakdown by category
• Dependency islands found
• Status information

Note: The EWI Fixer can be invoked separately via the spark-migration skill (which loads snowpark-api/dvp/dvp-ewi-fixer/SKILL.md as a bundled sub-skill). It uses the same database, so any changes will be reflected when you refresh the dashboard.

Step 5: Commit Changes to Git

After the dashboard is generated, commit the changes:

result = sma_api.git_commit("<workload_path>", """DVP SMA Dashboard: Generated interactive dashboard

Dashboard: sma-dashboard/
Modules: EWI Tracker, File Tracker, Dependency Tracker
Database: sma_storage.sqlite3""")

Verify branches:

result = sma_api.git_verify_branches("<workload_path>")

Script Options

Output Structure

sma-dashboard/
├── manifest.json               # Dashboard configuration
├── index.html                  # Main dashboard with side panel
├── server/
│   └── sma_server.py           # Standalone server script
├── assets/
│   └── styles.css              # Stylesheet
└── ewi-tracker/
    ├── ewi_tracker.html        # EWI Tracker view
    ├── file_tracker.html       # File Tracker view
    ├── dependency_tracker.html # Dependency Tracker view
    └── content/
        ├── files_*.html        # Detail pages per EWI
        └── file_ewis_*.html    # Detail pages per File

The folder is standalone - you can share it and run the server later with:

cd sma-dashboard
python3 server/sma_server.py .

Database Tables

issues

Main table storing EWI data from Issues.csv with status tracking.

artifact_dependency_inventory

Stores all artifact dependencies from ArtifactDependencyInventory.csv.

artifact_dependency_summary

Aggregated view of dependencies per file with island assignment.

artifact_dependency_graph

Stores edges for the dependency graph visualization.

Status Values

Database Access Module

A shared Python module provides programmatic access to sma_storage.sqlite3 so AI agents can query and update dashboard data conversationally, without needing the browser UI.

Module location: skills/spark-migration/snowpark-api/scripts/sma_api.py

All functions receive workload_path: str as their first parameter (the directory containing sma_storage.sqlite3).

Available Functions

All functions receive workload_path: str as first parameter.

Initialization

Read — EWI

Read — Files

Read — Dependencies

Write — Status

EWI Fixer

Reset

Overview

Final Summary

MANDATORY: After completing all steps (whether running standalone or invoked from the orchestrator), ALWAYS present this summary table:

SMA Dashboard Complete

┌───────────────────┬──────────┬──────────────────────────────────────────────────────┐
│ Step              │ Status   │ Details                                              │
├───────────────────┼──────────┼──────────────────────────────────────────────────────┤
│ SMA Dashboard     │ Done     │ Generated dashboard at sma-dashboard/                │
└───────────────────┴──────────┴──────────────────────────────────────────────────────┘

Output location: <output>/sma-dashboard/

Git branches:
• main — original code (unmodified)
• sma/migration-process — dashboard generation changes applied

Rules:

• Status is Done, Skipped, or Failed
• If dashboard generation failed, show Failed with brief error
• The git branches section uses sma_api.git_verify_branches() to confirm both branches exist