Go to file
ge c0d4ad668b upd 2023-04-26 16:46:36 +03:00
Pipfile init 2023-04-26 16:22:48 +03:00
Pipfile.lock init 2023-04-26 16:22:48 +03:00
README.md upd 2023-04-26 16:46:36 +03:00
cursed-archivebox-api.service upd 2023-04-26 16:46:36 +03:00
cursed_archivebox_api.py upd 2023-04-26 16:46:36 +03:00
requirements.txt init 2023-04-26 16:22:48 +03:00


Cursed API for ArchiveBox

ArchiveBox doesn't have a web API yet. This is shitty single-endpoint API to automate page archiving. It uses subprocess to run archivebox CLI. archivebox CLI runs in new thread to avoid blocking main thread.


You must deploy API to server which ArchieveBox runs.

Make working directory, create virtualenv and install dependencies:

apt install -y git python3-venv
mkdir -p /opt/cursed_archievebox_api && cd /opt/cursed_archievebox_api
python -m venv env
source env/bin/activate
pip install bottle gunicorn
git clone https://git.nxhs.cloud/ge/cursed-archivebox-api.git .

Start API. Set actual path to docker-compose.yml:

ARCHIVEBOX_BIN="docker compose -f /opt/archievebox/docker-compose.yml run archivebox" python cursed_archivebox_api.py

See also cursed-archivebox-api.service.

Install service:

cp cursed-archivebox-api.service /etc/systemd/system
systemctl enable --now cursed-archivebox-api.service

See logs:

journalctl -u cursed-archivebox-api.service -f


Variable Default
ARCHIVEBOX_BIN /usr/bin/archivebox (default for non-Docker installations)
CURSED_SERVER gunicorn See server backends

GET /add

Query parameters:

  • url. Resource URL
  • depth. Archive depth. Default: 0 (current page)
  • tag. List of comma separated tags e.g. my_tag, my_tag,another_one.

Example request:

curl -i 'http://localhost:9998/add?url=https://example.com&depth=0&tag=api,example'