orca-recipes
Health Uyari
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Basarisiz
- rm -rf — Recursive force deletion command in orca-docker-vm/docker-base-image.sh
Permissions Gecti
- Permissions — No dangerous permissions requested
Bu listing icin henuz AI raporu yok.
Copier template: isolated local Docker dev container per Orca ADE workspace, with Claude Code, git, gh/glab pre-authenticated via repo-scoped tokens. GitHub and GitLab (incl. self-managed).
🐳 orca-recipes
One-click isolated Docker environment for every Orca ADE workspace
English · Русский · 日本語 · 简体中文 · Português (BR) · Español · Français
✨ What it is
A recipe template for Orca ADE that creates an isolated environment in your local Docker for
every workspace, in one click.
- 🔐 Fine-grained git tokens — the token is scoped to a single repository
- 🤖 OAuth login for agents — log in once, it works in every workspace
- 🗝️ Secrets in the system keychain — macOS Keychain / Linux Secret Service, nothing ends up in the image
- 🐳 Docker inside the workspace —
docker run/build/composevia rootless Podman, no--privileged(optional)
🤖 Supported agents
| Agent | Status |
|---|---|
| Claude Code | ✅ Yes |
| cursor-agent | ✅ Yes |
| opencode (v2) | ✅ Yes |
| …others | 💡 PRs welcome |
🚀 Installation
Requirements: macOS or Linux, Docker (or OrbStack), jq, uv.
1. Render the template in the root of your project:
uvx copier copy gh:JargeZ/orca-recipes .
[!NOTE]
If you already had anorca.yaml, the template won't overwrite it — add the recipe yourself:environmentRecipes: - id: docker name: Local Docker checkoutMode: provisioned-root create: ./orca-docker-vm/docker-create.sh destroy: ./orca-docker-vm/docker-destroy.sh
2. Write your dev.Dockerfile — install everything your project needs for development.
The template creates a Python + uv example (template/dev.Dockerfile); adapt it
to your stack. The image requirements are in the comments at the top of the file: Debian/Ubuntu base
only, and any path outside the repo that dependency installation writes to must be owned by uid 1000.
3. Run the prepare script (safe to rerun):
./orca-docker-vm/prepare.sh
It asks which agents to set up (Claude Code, Cursor Agent, OpenCode), sets up the git token, builds the base image, logs the chosen agents in and runs a self-test.
Once everything is green, create a new workspace in the Orca UI and check that it comes up.
If something goes wrong, ask your agent to fix it: after rendering, all the recipe code lives in
your repository (orca-docker-vm/).
[!TIP]
Using Task? Addorca: ./orca-docker-vm/Taskfile.yamltoincludes
and runtask orca:setup.
To create a workspace from a script or an agent instead of the Orca UI, use./orca-docker-vm/workspace.sh add <name> and ./orca-docker-vm/workspace.sh rm <name>. Then use it
in the CLI as --environment <name>. See AGENTS.md.
🔄 Updating
Copier updates the template on top of your changes (three-way merge), so pulling a new version of
the recipe is one command:
./orca-docker-vm/update.sh # latest tag
./orca-docker-vm/update.sh --vcs-ref v1.2.0 # specific version
Updates never touch dev.Dockerfile or orca.yaml. Details are in AGENTS.md.
🛠️ Development
This project uses itself for development in Orca: clone the repo and run it through Orca in
dynamic environments.
The Taskfile is the entry point for tests and linters:
task test # render + shellcheck + orca doctor + update checks (fast, offline)
task test:e2e # full Docker run: build, create, SSH checks, destroy
🤝 Contributing
Pull requests that make it more universal (new agents, base images, git hosts) are very welcome!
📄 License
MIT: use it any way you like, keeping the copyright notice.
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi