QwenPaw with Docker Compose

What QwenPaw needs

Only figures QwenPaw's own documentation publishes, each linked to the page it is on. Where the project gives no number, this page does not invent one.

On top of this, Caddy (the reverse proxy in the file) is a single small container.

Which hardware QwenPaw has images for

Read from each image's manifest list in its registry (anonymous, read-only) when this page was built. An app counts for a CPU only if every one of its images lists it.

• x86-64 PC or NAS (linux/amd64): every image has a build.
• Raspberry Pi 4/5, 64-bit OS (linux/arm64): every image has a build.
• Raspberry Pi 3 or Zero 2 W, 32-bit OS (linux/arm/v7): no image for qwenpaw. No image for this hardware; the builder leaves the app out.

A manifest says an image exists for a CPU, not how well QwenPaw runs on it: a build for arm64 or arm/v7 can still be too slow or need more memory than the board has. The RAM and disk QwenPaw publishes are in the table above and, for a whole set of apps against a Raspberry Pi, on Will it run on my machine?. A tag like latest can move to a new digest after the date checked.

docker-compose.yml

This is exactly what the builder writes for QwenPaw alone on example.org: the official definition with its published ports removed and Caddy added in front. Every ${VARIABLE} is defined in the .env below.

# docker-compose.yml built at dankhost.com for: QwenPaw
# Each app block is its official compose definition. Only the names were prefixed with
# the app slug (services, containers, volumes, bind folders, .env variables) so any set of
# apps can share one file; other differences are listed under "Changed:" in each block.
# Only Caddy publishes ports. Secrets live in .env next to this file.

services:
  # ---- Caddy: the one reverse proxy, HTTPS certificates from Let's Encrypt ----
  # Source: https://caddyserver.com/docs/running (checked 2026-09-23)
  # Source: https://github.com/dani-garcia/vaultwarden/wiki/Using-Docker-Compose (checked 2026-09-23)
  # Caddyfile goes in ./caddy/Caddyfile. Fixed address 172.30.0.2 so apps can trust it exactly.
  caddy:
    image: "caddy:2"
    container_name: "caddy"
    restart: "unless-stopped"
    ports:
      - "80:80"
      - "443:443"
      - "443:443/udp"
    volumes:
      - "./caddy:/etc/caddy"
      - "caddy-data:/data"
      - "caddy-config:/config"
    networks:
      proxy:
        ipv4_address: "172.30.0.2"

  # ---- QwenPaw (qwenpaw.example.org) ----
  # Source: https://github.com/agentscope-ai/QwenPaw/blob/main/docker-compose.yml (checked 2026-10-01)
  # Source: https://github.com/agentscope-ai/QwenPaw/blob/main/website/public/docs/quickstart.en.md (checked 2026-10-01)
  # Source: https://github.com/agentscope-ai/QwenPaw/blob/main/website/public/docs/security.en.md (checked 2026-10-01)
  # Changed: Removed the documented '127.0.0.1:8088:8088' port, container_name and the version line; Caddy reaches qwenpaw:8088 on the proxy network. The three volumes keep their names, without the file's fixed name: lines.
  #          Turned on the web login that the file shows commented out: QWENPAW_AUTH_ENABLED is true, QWENPAW_AUTH_USERNAME admin, and QWENPAW_AUTH_PASSWORD a generated value in .env instead of the example. The security docs: authentication is disabled by default, and only requests from localhost skip it. Behind the proxy no request comes from localhost, so with the login off anyone who reached the address would get the console.
  qwenpaw:
    image: "agentscope/qwenpaw:latest"
    container_name: "qwenpaw"
    init: true
    restart: "always"
    environment:
      QWENPAW_AUTH_ENABLED: "true"
      QWENPAW_AUTH_USERNAME: "admin"
      QWENPAW_AUTH_PASSWORD: "${QWENPAW_AUTH_PASSWORD}"
    volumes:
      - "qwenpaw-data:/app/working"
      - "qwenpaw-secrets:/app/working.secret"
      - "qwenpaw-backups:/app/working.backups"
    networks:
      proxy: {}

networks:
  # Pinned so the trusted-proxy address given to each app (172.30.0.2) is exact.
  # If "docker compose up" reports "Pool overlaps", change 172.30.0 here, in Caddy's
  # ipv4_address and in every trusted-proxy value to a free 172.x.0 range.
  proxy:
    ipam:
      config:
        - subnet: "172.30.0.0/24"
          ip_range: "172.30.0.128/25"
          gateway: "172.30.0.1"

volumes:
  caddy-data:
  caddy-config:
  qwenpaw-data:
  qwenpaw-secrets:
  qwenpaw-backups:

Caddyfile save as caddy/Caddyfile

Caddy gets and renews the HTTPS certificate for qwenpaw.example.org by itself once the DNS record points at your server and ports 80 and 443 reach it.

# Caddyfile built at dankhost.com. Save it as ./caddy/Caddyfile next to docker-compose.yml.
# Caddy gets and renews a Let's Encrypt certificate for each name below by itself,
# once the DNS records point at this server and ports 80 and 443 reach it.

# QwenPaw
qwenpaw.example.org {
	reverse_proxy qwenpaw:8088
}

Already running nginx, Nginx Proxy Manager, Caddy or Traefik? The reverse-proxy settings QwenPaw needs, for all four, each cited to QwenPaw's own docs.

.env

The secret is left blank here on purpose: a password printed on a public page is the same for everyone who copies it. The builder fills it with random letters and digits made in your browser.

# .env for QwenPaw, from dankhost.com/app/qwenpaw/. Fill the blank secret before starting.
# Keep this file private (chmod 600).

# QwenPaw
# 24 random letters and digits: the builder fills this in your browser
QWENPAW_AUTH_PASSWORD=

Changes from the official file

Besides prefixing every service, container, volume, folder and variable with qwenpaw (so any set of apps can share one file), these are all the differences from what the project documents:

• Removed the documented '127.0.0.1:8088:8088' port, container_name and the version line; Caddy reaches qwenpaw:8088 on the proxy network. The three volumes keep their names, without the file's fixed name: lines.
• Turned on the web login that the file shows commented out: QWENPAW_AUTH_ENABLED is true, QWENPAW_AUTH_USERNAME admin, and QWENPAW_AUTH_PASSWORD a generated value in .env instead of the example. The security docs: authentication is disabled by default, and only requests from localhost skip it. Behind the proxy no request comes from localhost, so with the login off anyone who reached the address would get the console.

Notes for QwenPaw behind the proxy

• Log in at https://qwenpaw.example.org as admin with QWENPAW_AUTH_PASSWORD from .env: the security docs' auto-registration creates that single admin account at startup. A login lasts 7 days.
• Add a model in the console under Settings > Models (an API key, or the address of a local model server). Model settings and API keys are kept in the qwenpaw-secrets volume.
• QwenPaw is an agent that runs tools and shell commands inside its container; its security docs describe the tool guard and file guard that limit them.
• The image tag is 'latest', the stable release ('pre' is the pre-release).

Back it up

QwenPaw's Backup & Restore page: create a full backup in the console (written to the qwenpaw-backups volume), export it to keep a copy elsewhere, and restore from the same page, completely or agent by agent. Copying the three volumes below covers the same data.

Below is the BACKUP.md the builder writes next to the compose file above: which of its folders and volumes hold data, what to copy with which container stopped, and a restore check. Pick more apps in the builder and it covers them all in one file.

# BACKUP.md built at dankhost.com for: QwenPaw

What to copy from the docker-compose.yml built with this file, the database dump each project documents (rewritten for the service names in that file), what to copy with which container stopped, and how to check the result. Run every command in the folder that holds docker-compose.yml.

The folders are owned by the containers' users: put sudo in front of tar if it says "Permission denied". Named volumes live under /var/lib/docker/volumes, not next to the compose file, so they are copied through a throwaway ubuntu container (Docker's documented --volumes-from method). tar makes a full copy each run; for large folders an incremental tool (restic, borg, rsync) saves time and space.

Never run "docker compose down -v" on this file: it deletes the named volumes of every app in it.

## Before you start

    BACKUP=/mnt/backup/selfhost   # change this: a disk or machine other than this server's system disk
    mkdir -p "$BACKUP"

## Every time: the three stack files

    tar czf "$BACKUP/stack-files.tar.gz" docker-compose.yml .env caddy

.env holds the database passwords and secret keys the apps were set up with; keep this archive as private as the data.

## Caddy (the reverse proxy)

Caddy's docs: 'The data directory must not be treated as a cache.' It holds the TLS certificates and their private keys; without it Caddy has to ask Let's Encrypt for every certificate again.

- caddy-data (named volume, /data in caddy)
  BACK UP: TLS certificates, private keys, OCSP staples.
- caddy-config (named volume, /config in caddy)
  SKIP: the last active configuration, kept for 'caddy run --resume', which this setup does not use: it starts from the Caddyfile.
- ./caddy (/etc/caddy in caddy)
  IN stack-files.tar.gz: the Caddyfile, copied with the stack files above.

Copy (Caddy keeps serving while you copy):

    docker run --rm --volumes-from caddy -v "$BACKUP":/backup ubuntu tar cvf /backup/caddy-data.tar /data

Restore check:

    tar tf "$BACKUP/caddy-data.tar" > /dev/null && echo OK
    docker run --rm --volumes-from caddy -v "$BACKUP":/backup ubuntu bash -c "cd /data && tar xvf /backup/caddy-data.tar --strip 1"

Sources:
- Caddy docs: Conventions, Data directory: https://caddyserver.com/docs/conventions#data-directory (checked 2026-09-24)
- Docker docs: Volumes, Back up, restore, or migrate data volumes: https://docs.docker.com/engine/storage/volumes/#back-up-restore-or-migrate-data-volumes (checked 2026-09-24)

## QwenPaw

QwenPaw's Backup & Restore page: create a full backup in the console (written to the qwenpaw-backups volume), export it to keep a copy elsewhere, and restore from the same page, completely or agent by agent. Copying the three volumes below covers the same data.

What this compose mounts:
- qwenpaw-data (named volume, /app/working in qwenpaw)
  BACK UP: config, memory, skills and the agents' workspaces.
- qwenpaw-secrets (named volume, /app/working.secret in qwenpaw)
  BACK UP: model provider settings and API keys; keep this copy private.
- qwenpaw-backups (named volume, /app/working.backups in qwenpaw)
  BACK UP: the backup archives made in the console.

1. Stop qwenpaw. QwenPaw writes memory and config files as the agent works; stopping it keeps the three volumes in step.

    docker compose stop qwenpaw

2. Copy while qwenpaw is stopped:

    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu tar cvf /backup/qwenpaw-data.tar /app/working
    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu tar cvf /backup/qwenpaw-secrets.tar /app/working.secret
    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu tar cvf /backup/qwenpaw-backups.tar /app/working.backups

3. Start it again:

    docker compose start qwenpaw

Restore check:

    tar tf "$BACKUP/qwenpaw-data.tar" > /dev/null && echo OK
    tar tf "$BACKUP/qwenpaw-secrets.tar" > /dev/null && echo OK
    tar tf "$BACKUP/qwenpaw-backups.tar" > /dev/null && echo OK

Put a copy back (with qwenpaw stopped):

    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu bash -c "cd /app/working && tar xvf /backup/qwenpaw-data.tar --strip 2"
    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu bash -c "cd /app/working.secret && tar xvf /backup/qwenpaw-secrets.tar --strip 2"
    docker run --rm --volumes-from qwenpaw -v "$BACKUP":/backup ubuntu bash -c "cd /app/working.backups && tar xvf /backup/qwenpaw-backups.tar --strip 2"

Put the three volumes back and run docker compose up -d. Or import an exported backup in the console and restore it there; the docs recommend ticking "Create a pre-restore backup first".

Sources:
- QwenPaw docs: Backup & Restore: https://github.com/agentscope-ai/QwenPaw/blob/main/website/public/docs/backup.en.md (checked 2026-10-01)
- QwenPaw docs: Quick start, Option 3: Docker (image tags, the three volumes): https://github.com/agentscope-ai/QwenPaw/blob/main/website/public/docs/quickstart.en.md (checked 2026-10-01)
- Docker docs: Volumes, Back up, restore, or migrate data volumes: https://docs.docker.com/engine/storage/volumes/#back-up-restore-or-migrate-data-volumes (checked 2026-10-01)

Update it

What docker compose pull does depends only on how each image line in the compose file above is tagged. For QwenPaw as this page writes it:

What QwenPaw's docs say about updating

QwenPaw's FAQ, "How to update QwenPaw": "If using Docker, pull the latest image and restart the container". Its example pulls agentscope/qwenpaw:latest and runs it again with the same three volumes:

docker pull agentscope/qwenpaw:latest

• docker compose pull then docker compose up -d, below, is the compose form of these steps: the new container gets the same qwenpaw-data (config, memory and skills), qwenpaw-secrets (model settings and API keys) and qwenpaw-backups volumes.
• The FAQ also says data from the older CoPaw name carries over: "No extra migration is needed".

Source: QwenPaw docs: FAQ, How to update QwenPaw checked 2026-10-01

The steps

1. Back up first, as in Back it up above. An update can migrate QwenPaw's data on its first start; the copy is the way back.
2. Download the new images. Running containers are not touched yet:

   docker compose pull

3. Recreate the containers whose image changed. Volumes and bind-mounted folders stay as they are:

   docker compose up -d

4. Watch QwenPaw start (Ctrl+C stops following the log, not the app):

   docker compose logs -f qwenpaw

5. Optional: remove the old images the pull left without a tag. This covers every stack on the machine, not only this one:

   docker image prune

Docker's reference for these commands: Docker docs: docker compose pull checked 2026-09-30; Docker docs: docker compose up (recreates changed containers, keeps volumes) checked 2026-09-30; Docker docs: docker image prune checked 2026-09-30

Sources

Every page below was fetched on the date shown.

• QwenPaw on GitHub (repository description) checked 2026-10-01 (what it does)
• QwenPaw: docker-compose.yml in the repository checked 2026-10-01 (compose definition and proxy settings)
• QwenPaw docs: Quick start, Option 3: Docker (image tags, the three volumes) checked 2026-10-01 (compose definition and proxy settings, backup plan)
• QwenPaw docs: Security, Web Authentication (QWENPAW_AUTH_ENABLED, auto-registration, localhost bypass) checked 2026-10-01 (compose definition and proxy settings)
• Caddy docs: Keep Caddy running (Docker Compose section) checked 2026-09-23 (Caddy)
• Vaultwarden wiki: Caddy with HTTP challenge (uses the caddy:2 image) checked 2026-09-23 (Caddy)
• QwenPaw docs: Backup & Restore checked 2026-10-01 (backup plan)
• Docker docs: Volumes, Back up, restore, or migrate data volumes checked 2026-10-01 (backup plan)
• Caddy docs: Conventions, Data directory checked 2026-09-24 (backup plan)
• QwenPaw docs: FAQ, How to update QwenPaw checked 2026-10-01 (update notes)
• Docker docs: docker compose pull checked 2026-09-30 (update steps)
• Docker docs: docker compose up (recreates changed containers, keeps volumes) checked 2026-09-30 (update steps)
• Docker docs: docker image prune checked 2026-09-30 (update steps)

Add QwenPaw to the builder All seventy apps