Ora (ora.pm) is shutting down on Saturday, October 31, 2026. Its JSON and CSV exports only contain links to the files attached to your tasks (images, PDFs, etc.). Those files are stored on Amazon S3 and will disappear along with Ora.

This tool downloads those files to your computer and sorts them into folders: one per project, and one per task.

⏰ Do it soon. Once Ora shuts down, the files can no longer be downloaded.

Allow about 15 minutes the first time. Nothing to buy, no account needed.

A Mac connected to the Internet.

The file ora_attachments.py (this tool).

Your Ora exports: one .json file per project (see step 1). The .csv export is only a last resort: it contains almost no attachments.

You do not need the Ora app.

Step 1 — Export your projects from Ora and prepare your folders

First, export each project as JSON, while Ora still works:

At the very bottom of the window, under Entire Project, click Export as JSON (circled in red below). Export as CSV: isn't the correct format to locate attachments, as it contains almost none.

A .json file named after the project is downloaded, usually into your Downloads folder.

⚠️ The Ora - Attachments folder must not be inside the Ora - Exports folder. The tool refuses to write into the exports folder, to protect your exports.

Terminal is a window where you type commands.

In Terminal, type exactly this, then press Return:

python3 --version

If you see Python 3. followed by a number, go to step 4.

If a window offers to install the "command line developer tools", click Install, accept, and wait until it finishes (about 10 minutes). Then type python3 --version again and press Return.

Step 4 — Start the download

You will build a command by dragging folders into the Terminal window, so you never have to type a path. Put Finder and Terminal side by side.

The line should look like this (with your user name instead of marie):

python3 /Users/marie/Documents/ora_attachments.py export --exports /Users/marie/Documents/Ora\ -\ Exports --out /Users/marie/Documents/Ora\ -\ Attachments --pictures

The \ before spaces are normal: Terminal adds them by itself when you drag a folder.

Lines scroll by, one per file, for example [12/262] ok …. Keep the window open until the word Summary appears. It takes a few minutes.

Project ok skipped failed ... TOTAL 265 0 1

ok: files downloaded. 🎉

skipped: files already there from a previous run, so they were skipped. That's normal.

failed: files that could not be downloaded. The list is shown below the table.

If some files failed, run the same command again: press the ↑ arrow key (the command comes back), then Return. The tool only fetches what is missing.

A missing_remote failure (code 403 or 404) means the file no longer exists at Ora. Running again won't change that.

Type python3 and a space, drag ora_attachments.py, type verify --out and a space, drag the Ora - Attachments folder, then press Return.

You should see … file(s) verified without problems. Files marked missing_remote are also listed as problems, which is normal.

#My project/ #489 - Task title/ 2457898 - image.png ← the file, filed under its task _no-task/ ← files whose task is not in the export _pictures/ ← project pictures and member photos manifest.csv ← the list of all files (opens in Excel or Numbers) external-links.csv ← Google Drive, Figma… links for you to save yourself

Troubleshooting

What you see What to do

command not found: python3 Do step 3 again.

can't open file … ora_attachments.py The file is not where you think. Redo step 4, dragging the file instead of typing its name.

Exports not found Redo step 4, dragging the Ora - Exports folder.

is inside the exports folder (read-only) Your output folder is inside the exports folder. Move it next to it (step 1).

TLS certificate error Open the Applications folder, then the Python 3.x folder, and double-click Install Certificates.command. Then run the command again.

The connection dropped, or you closed the window Run the same command again: nothing is lost, the tool resumes where it stopped.

WARNING: … read from CSV The .json file for that project is missing. The CSV contains almost no attachments (see below). Add the .json if you can.

Windows and Linux

Windows: install Python from python.org, ticking "Add python.exe to PATH". Then open Command Prompt and follow the same steps, typing python instead of python3. Pick an output folder with a short path, such as C:\Ora, because Windows limits path length.

Linux: Python 3 is almost always already installed. Follow the same steps in a terminal.

Reference

Python 3.9 or later. Standard library only: nothing to install.

Download (resumes where it stopped if run again)

python3 ora_attachments.py export --exports "<folder or file>" --out "<folder>" [--source json|csv|auto] [--pictures] [--dry-run]

Sort files you already downloaded some other way, without downloading anything

python3 ora_attachments.py relink --exports "<folder or file>" --files "<folder>" --out "<folder>"

Check the files (presence, size, SHA-256). Non-zero exit code if there is a problem.

python3 ora_attachments.py verify --out "<folder>"

--source auto (default) uses a project's JSON if present, otherwise its CSV.

--pictures also saves project pictures and member photos into _pictures/.

--dry-run shows the plan without writing or downloading anything.

relink finds each file by its S3 key name (last segment of the URL), or by its SHA-256 if --files contains a manifest.json. Files are copied, never moved.

Manifest statuses: ok, skipped (already present and verified on a rerun), failed, size_mismatch (size differs from Ora's fileSize) and missing_remote (403 or 404: the file no longer exists on S3).

The CSV has no attachment column and only contains active tasks. The only files that can be found in it are the images embedded in task descriptions. In one real export, that was 15 files, versus 246 attachments in the JSON. The tool prints a warning whenever a project is read from CSV. Use the JSON whenever possible.

It never calls the Ora API (api.ora.pm, GraphQL, WebSocket) and never logs in to any account. It only makes HTTP GET requests to the public URLs found in the exports.

It never writes into the exports folder.

It does not download external links (Google Drive, Figma…): it lists them in external-links.csv.

Privacy

S3 links are public, and JSON exports contain the members' email addresses. Sharing the script is fine. Do not share your exports or your Ora - Attachments folder.

python3 -m unittest discover -s tests

The tests use fictitious fixtures and a local fake S3 server. No real data is stored in this repository.