viwo.example.yml

# Example VIWO project configuration file
# Place this file at the root of your repository as viwo.yml or viwo.yaml

# Post-install hooks run on the host after git worktree creation
# and before the containerized agent runtime starts.
# Commands execute in the worktree directory.
# If any command fails, session initialization will fail.
postInstall:
  # Install dependencies
  - npm install

  # Build the project
  - npm run build

  # Copy environment file
  - cp .env.example .env

  # Any other setup commands your project needs
  # - docker-compose up -d
  # - npx prisma generate

# Pre-agent commands run inside the container before the agent starts.
# Useful for installing dependencies or building the project.
#
# preAgent:
#   - bun install
#   - bun run build

# Project Dockerfile that extends the viwo base image. The file's first
# non-comment line must be `FROM overseedai/viwo-claude-code:<version>`
# matching the viwo CLI's pinned base. Built once per content hash and
# tagged `viwo-derived:<hash>`; cached for subsequent sessions.
#
# Use this when you need system packages or language runtimes the base
# image doesn't ship — installing them in `preAgent` is slow and runs as
# the non-root `claude` user, so it can't `apt-get install`.
#
# A working `viwo.Dockerfile` for this repo is checked in alongside this
# example — copy this file to `viwo.yml` to opt in.
#
# dockerfile: ./viwo.Dockerfile

# Custom bind mounts exposed inside the agent container.
# Host paths may be absolute, use `~` for your home directory, or be
# relative to the repository root. Container paths must be absolute.
# Append `:ro` (or set `readonly: true`) for read-only mounts.
#
# binds:
#   # String form (Docker-style)
#   - ~/.cache/huggingface:/root/.cache/huggingface
#   - ./shared-data:/shared:ro
#
#   # Object form
#   - source: ~/models
#     target: /models
#     readonly: true