本文へ移動
cccskills
無料GitHub で公開

daytona

Daytona CLI setup, sandbox debugging, keep a sandbox alive, secrets volume, snapshot refresh. Use when Daytona itself is the problem, not for running tests (see run-tests).

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.6 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Daytona

Use this skill for Daytona infrastructure, not test placement.

Setup

Install the Daytona CLI with brew install daytonaio/cli/daytona, then authenticate:

daytona login

Which identity the CLI uses

  • Environment key. The CLI uses DAYTONA_API_KEY only when DAYTONA_API_URL=https://app.daytona.io/api is also set; with the key alone it silently falls back to the saved profile. Scope the team key (Infisical dev, /openwork-ops) to one command: DAYTONA_API_URL=https://app.daytona.io/api DAYTONA_API_KEY="$(infisical secrets get DAYTONA_API_KEY --env dev --path /openwork-ops --plain --silent)" <command>.
  • Saved profile. Otherwise the active profile in $DAYTONA_CONFIG_DIR/config.json (default: macOS ~/Library/Application Support/daytona/, Linux ~/.config/daytona/, Windows %APPDATA%\daytona\). Point DAYTONA_CONFIG_DIR at an empty directory to test the not-logged-in path without touching the real profile.
  • Never run daytona login --api-key on someone's machine for a one-off task. It overwrites their profile, and with it their browser login, for every tool. A key printed as *not found* (Infisical CLI 0.28.x exits 0 on a missing secret) is not a key.
  • Personal-organization trap. A browser login may default to the person's personal organization: a 10 GiB total memory cap and none of the team's warm snapshots, so multi-sandbox previews fail with a memory-limit error. pnpm world plan <world> --place daytona prints the identity and organization in use and warns on a personal one.
  • Version warning. "Daytona CLI is on vX and API is on vY" on every call is a warning, usually not the cause of a failure; fix it with brew upgrade daytonaio/cli/daytona.

To run tests, just use pnpm evals:e2e <slug> — it picks Daytona automatically when the CLI is authenticated. --local forces local; --daytona requires Daytona and fails when it is unavailable.

For a PR preview that should open inside Codex and support named scenarios, use preview-my-work. It composes the same maintained world and Daytona primitives with scoped ownership and teardown.

Long-lived manual Electron sandbox

Run bash .devcontainer/test-on-daytona.sh <ref>. The maintained helper uses the reusable desktop snapshot, starts XFCE/noVNC, Vite, and Electron, and prints the sandbox and preview URLs. Keep that sandbox for exploration or debugging instead of reproducing its provisioning commands.

Long-lived manual server sandbox

Run bash .devcontainer/test-server-on-daytona.sh <ref>. It starts the separate MySQL, Den API, and Den Web sandbox and prints public URLs for a desktop sandbox to consume. Use this helper rather than assembling the server manually.

Debugging

Inspect /tmp/start-vnc.log, /tmp/vite.log, /tmp/electron.log, and /tmp/den-api.log in the relevant sandbox. Electron CDP is port 9825; get CDP and noVNC URLs with:

daytona preview-url <sandbox> -p 9825
daytona preview-url <sandbox> -p 6080

Secrets

bash .devcontainer/setup-daytona-secrets-volume.sh <local-env> <name>.env

Electron sandboxes mount the reusable secrets volume at /daytona-secrets.

Snapshot refresh

bash .devcontainer/create-daytona-openwork-snapshot.sh
bash .devcontainer/create-daytona-openwork-server-snapshot.sh

Teardown

daytona delete <sandbox>

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Add a new feature, put work behind a feature flag, roll something out (per organization, for everyone, cloud or self-hosted), revert or kill a feature, or make a big change to existing behavior such as a new engine. Use BEFORE writing the feature code, and when finishing or removing a rollout.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

Local OpenWork Electron browser automation with CDP. Use when driving a local Electron dev app, browser_list, browser_snapshot, browser_eval, composer automation, or local UI smoke tests.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

Flag customer, prospect, partner, or outside-person identities in the diff. Reported in the Warden security summary.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

Connect the OpenWork MCP Gateway (https://api.openworklabs.com/mcp/agent) to Claude Code, Codex, Gemini CLI, Cursor, VS Code, Claude Desktop, or ChatGPT so the agent can use the user's OpenWork organization skills, plugins, and connections.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

Screen a pull request from an outside contributor for hidden, obfuscated, or supply-chain-risky changes before any of its code runs.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

Create an OpenCode plugin for OpenWork. Scaffolds the plugin file with the correct API shape, tool definitions, and hook registration. Use when the user asks to 'create a plugin', 'write a plugin', or 'make a plugin that does X'.

日本語の概要は準備中です。原文の説明を表示しています。

different-ai/openwork2.4万2026年10月11日 更新

different-ai のスキルをすべて見る

このスキルの問題を報告する