slack-exporter

A Slack bot and standalone script for exporting messages and file attachments from public and private channels, using Slack's new Conversations API.

A similar service is provided by Slack for workspace admins at https://my.slack.com/services/export (where my can be replaced with your full workspace name to refer to a workspace different than your default). However, it can only access public channels, while slack-exporter can retrieve data from any channel accessible to your user account.

Authentication with Slack

There are two ways to use slack-exporter (detailed below). Both require a Slack API token to be able to communicate with your workspace.

1. Visit https://api.slack.com/apps/ and sign in to your workspace.
2. Click Create New App. If prompted to select "how you'd like to configure your app's scopes", choose the App Manifest option. You can configure the app manually instead, but you will be prompted to enter an app name and additional steps to set up permissions instead of the single step below. Once created, select your workspace.
3. You should then be prompted for an app manifest. Paste the contents of the slack.yaml file (in the root of this repo) into the YAML box.
4. Select Install App in the side bar, then Install to Workspace (or Reinstall to Workspace if you have done this previously) and accept at the prompt.
5. Copy the OAuth Access Token (which will generally start with xoxp for user-level permissions and may be located in a section like "OAuth & Permissions" in the sidebar).

Usage

As a standalone script

exporter.py can create an archive of all conversation history in your workspace which is accessible to your user account.

1. Either add

   SLACK_USER_TOKEN = xoxp-xxxxxxxxxxxxx...
   

   to a file named .env in the same directory as exporter.py, or run the following in your shell (replacing the value with the user token you obtained in the Authentication with Slack section above).

   export SLACK_USER_TOKEN=xoxp-xxxxxxxxxxxxx...

2. If you cloned this repo, make sure that dependencies are installed by running pip install -r requirements.txt in the repo root directory.

3. Run python exporter.py --help to view the available export options. You can test that access to Slack is working by listing available conversations: python exporter.py --lc.

As a Slack bot

bot.py is a Slack bot that responds to "slash commands" in Slack channels (e.g., /export-channel). To connect the bot to the Slack app generated in Authentication with Slack, create a file named .env in the root directory of this repo, and add the following line:

SLACK_USER_TOKEN = xoxp-xxxxxxxxxxxxx...

Save this file and run the Flask application in bot.py such that the application is exposed to the Internet. This can be done via a web server (e.g., Heroku), as well as via the ngrok service, which assigns your localhost server a public URL.

To use the ngrok method:

1. Download the appropriate binary.

2. Run python bot.py

3. Run the ngrok binary with path/to/ngrok http 5000, where 5000 is the port on which the Flask application (step 2) is running. Copy the forwarding HTTPS address provided.

4. Create the following slash commands will be created (one for each applicable Flask route in bot.py):

   Command	Request URL	Arguments	Example Usage
   /export-channel	https://[host_url]/slack/export-channel	json | text	/export-channel text
   /export-replies	https://[host_url]/slack/export-replies	json | text	/export-replies json

   To do this, uncomment the slash-commands section in slack.yaml and replace YOUR_HOST_URL_HERE with something like https://xxxxxxxxxxxx.ngrok.io (if using ngrok). Then navigate back to OAuth & Permissions and click (Re)install to Workspace to add these slash commands to the workspace (ensure the OAuth token in your .env file is still correct).

CSV export

Use --csv with the exporter to output in CSV format instead of plain text:

python exporter.py --csv -c -o ./exports

Each channel is written to a separate .csv file with columns: timestamp, user, text, thread_ts, reply_count, and media_data. Attached files and images are downloaded and embedded as base64 data URIs in the media_data column.

To convert existing JSON exports to CSV after the fact, you can use converter_to_csv.py:

python converter_to_csv.py exports/slack_export_2024-01-01_120000

Pass --user-list with a user list JSON (from python exporter.py --lu --json -o ./exports) to resolve user IDs to display names.

Testing

Warning

The test suite creates channels, posts messages, archives/unarchives channels, and performs other write operations in your Slack workspace. Only run these tests against a dedicated test workspace. Do not run them against a production workspace.

Setup

1. Add write scopes to your Slack app manifest. In slack.yaml (or in your app settings at https://api.slack.com/apps/), paste the following scopes under oauth_config.scopes.user alongside the existing read scopes:

   - channels:write
   - chat:write
   - groups:write
   - im:write
   - mpim:write

   After updating scopes, click Reinstall to Workspace (under OAuth & Permissions) and update the token in your .env file if it has changed.

2. Install test dependencies:

   pip install -r requirements-test.txt

3. Run the tests:

   pytest tests/ -v

   Before any tests execute you will be shown the workspace name and prompted to confirm. To skip the interactive prompt (e.g., in CI), pass --force:

   pytest tests/ --force -v

What the tests do

File	Scope	Description
test_parsing.py	Unit	Parsing helpers, name lookups, rate-limit retry, download (mocked HTTP)
test_api_integration.py	Integration	Channel list/history, user list, replies, time-range filters, archived-channel access against the real Slack API
test_bot.py	Unit	Flask route logic with mocked exporter functions, file download/cleanup

The integration tests create temporary channels whose names start with _test_ (e.g., _test_pub_a1b2c3d4). These are archived automatically when the test session ends. Deletion is restricted to the admin scope, which requires an Enterprise plan, so we can't fully clean up automatically.

Note: Integration tests make real API calls and may be slow if rate-limited. The concurrency group ensures only one test run executes at a time to avoid channel-name collisions or conflicting state.

Contributors

• Seb Seager - original author
• Gregor Kobilarov - CSV export functionality and media file support

License

This software is available under the GPL.