Free local evaluation · version 1.3.1
Your exports. Your computer.
A browser workspace powered by a small local Python service. No account, payment or expiry. No extension installation is required.
Prerequisites: Python 3.10 or newer, a modern browser and permission to run local software. Python is not bundled. Your operating system may request permission to run downloaded scripts. Follow your organisation’s software policy.
1. Extract the complete ZIP
Keep all files inside CRM_Migration_Checker_Evaluation. Do not run files from within the ZIP preview.
2. Check Python and start
macOS or Linux: open Terminal in the extracted folder, run python3 --version, then python3 launch.py. On macOS, Start_Mac.command offers the same launch if your system permits it.
Windows: open a terminal in the folder, run py -3 --version, then py -3 launch.py. Alternatively, run Start_Windows.cmd.
If Python is missing or older than 3.10, install a current supported Python from the official Python website or ask your IT administrator. If macOS selects an older built-in Python despite a newer installation, use the newer interpreter’s full path to run launch.py.
3. Open the local workspace
The launcher verifies packaged file hashes, selects a free local port and opens your browser. If it does not open automatically, copy the printed http://127.0.0.1:…/workspace/ URL. Keep the terminal open while working.
4. Run the fictional example
- Select “Load fictional exports”.
- Review the retained source ID, selected fields and comparison rules.
- Confirm the rules and run the comparison.
- Review findings and coverage. Download HTML, CSV and JSON evidence.
- Change a rule or add an explicitly accepted value change, then rerun. Previous runs remain available in this tab, up to three runs.
For your own files, choose only exports you are authorised to process. Both files must be comma-separated UTF-8 CSV. Start with a small subset and a stable retained record ID.
5. Save and stop
Save required downloads before clearing or closing the tab. Press Control-C in the launcher terminal to stop; active requests finish cleanup before exit. Closing the browser alone does not stop the Python service.
Limits and privacy
Per file: 5,000 data rows, 10 MiB, 128 columns, 250,000 data cells and 10,000 characters per cell. One to five selected fields. Up to 10,000 findings. Value maps: 1 MiB total and up to 1,000 entries total. Evidence: 16 MiB; individual downloads and the combined browser response: 32 MiB. All limits apply together; oversized work fails rather than truncate.
Processing uses local memory and temporary plaintext files. Temporary working copies are removed after successful or failed requests complete normally. Forced process termination or a device crash can leave temporary files. Original inputs and saved downloads are not erased. The app does not encrypt them: use your own storage, access and retention controls. The workspace makes no external network requests and uses no telemetry or external assets.
The report covers selected rules in the supplied snapshots. Zero findings is not certification of a complete migration. Attachments, effective permissions, live workflows, export completeness and business meaning need separate checks.
Troubleshooting and integrity
Run python3 launch.py --check (Windows: py -3 launch.py --check) to check file integrity. The SHA-256 manifest detects changes or corruption; it is not a digital signature. Re-extract a fresh package if files are missing or changed.
--no-browser prints the local URL without opening a tab. --port 8773 selects a specific local port. If it is unavailable, omit that option. No administrator rights or third-party Python packages are required by this app.