Agent skill

OpenWork PR Preview Manager

by different-ai in different-ai/openwork

Boots, reopens, updates or resets disposable OpenWork preview environments for a pull request, locally or on a private Daytona or Freestyle host.

Custom licenceAuto-check passedDevOps & Cloud

Install OpenWork PR Preview Manager

skills CLI
$ npx skills add different-ai/openwork --skill preview-my-work -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install different-ai/openwork preview-my-work --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/different-ai/openwork.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.opencode/skills/preview-my-work .claude/skills/preview-my-work && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
preview-my-work
GitHub stars
24k
Token cost
~5.9k tokens
SKILL.md length
3,093 words
Files
2 (incl. scripts)
Skills in repo
33
Repo updated
First seen
Licence
Custom licence

At a glance

Boots, reopens, updates or resets disposable OpenWork preview environments for a pull request, locally or on a private Daytona or Freestyle host.

  • Works in 3 steps: pnpm world help --json: each target's… → pnpm world plan --place --stage : checks… → pnpm world up ... from an example. It…
  • Booting a preview of an OpenWork pull request for manual testing
  • SKILL.md covers Agent quick path, Choose a preview, Saved web evidence checkpoints and Start and open, plus 2 more sections
  • Runs Python scripts from its folder; calls pnpm; needs DAYTONA_API_KEY and FREESTYLE_API_KEY

What it does

This skill drives the repository's world lifecycle to give a pull request a running preview for hands-on testing. The available worlds are preview-desktop for the app only, preview-den for Den only, preview-full for Den plus the desktop app, and preview-app-web, and each can run locally, on a private Daytona URL or on Freestyle. The environments are disposable test setups, not production or an installed desktop profile, and the agent is told not to touch another world or an existing test sandbox.

A three-step quick path does the work. pnpm world help lists each target's seeds, sources, required logins and ready-to-run commands, pnpm world plan checks the requirements without creating anything and exits with the fix when one is unmet, and pnpm world up starts the preview after checking again. pnpm world outputs reports whether anything exists to reopen. Remote previews default to the current origin/dev commit, and a bundled update-preview.py script is included for refreshing a preview.

Credentials are loaded inside a command substitution for the single command that needs them and are never printed. Daytona needs both DAYTONA_API_URL and DAYTONA_API_KEY because its CLI silently ignores the key without the URL, and Freestyle uses FREESTYLE_API_KEY. The skill warns against running daytona login with an API key, since that replaces the person's CLI login on the machine.

When your agent uses it

  • Booting a preview of an OpenWork pull request for manual testing
  • Reopening or updating an existing preview after new commits
  • Choosing between desktop-only, Den-only and full previews
  • Running a preview on a private Daytona URL or on Freestyle instead of locally

Example prompts

  • “Spin up a full preview with Den and the desktop app for this branch on Daytona.”
  • “Reopen the existing preview for my PR and update it to the latest commit.”
  • “Check what the preview-den world needs before I start it, without creating anything.”
  • “Reset the preview-desktop world in this worktree and boot it locally.”

Requirements

  • pnpm and a checkout of the OpenWork repository
  • A Daytona API URL and key, or a Freestyle API key, for remote previews
  • Infisical CLI access to fetch those keys, if you use the team setup

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. pnpm world help --json: each target's seeds and sources, what an
  2. pnpm world plan --place --stage : checks those
  3. pnpm world up ... from an example. It checks the requirements again before

What it can do on your machine

Read from SKILL.md and the folder at commit 4ccc9e9. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • pnpm

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • daytona.io

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • DAYTONA_API_KEY
    • FREESTYLE_API_KEY

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

OpenWork PR Preview Manager loads about 5.9k tokens when it runs. Until then it costs about 69 tokens; SKILL.md has 3,093 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~69
When it runs · the whole SKILL.md, loaded when a task matches
~5.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 3,093 words (~5,906 tokens).

“Use the repository's world lifecycle. These are disposable test environments, not production or a user's installed desktop profile. Do not touch another world or an existing test sandbox. Run from the requested worktree.”

— opening of SKILL.md by different-ai, Custom licence
name
preview-my-work

Read the full SKILL.md on GitHub

Files

SKILL.md and 1 other file (scripts) in .opencode/skills/preview-my-work of different-ai/openwork.

  • SKILL.md
  • scripts/update-preview.py

Open the folder on GitHubat commit 4ccc9e9

Compare with similar skills

OpenWork PR Preview Manager next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

OpenWork PR Preview Manager compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
OpenWork PR Preview Manager this skilldifferent-ai/openwork24k—~5.9kAutomated safety check: PassCustom licence
Desktop Release Sync LandingEynzof/Hermes-CN-Desktop1.8k—~1.3kAutomated safety check: PassCustom licence
Kouchou AI Developmentdigitaldemocracy2030/kouchou-ai171—~540Automated safety check: NotesAGPL-3.0
Deploy Release Testvercel/next.js143k—~1.2kAutomated safety check: PassMIT
Svelte Deploymentspences10/svelte-claude-skills218—~380Automated safety check: PassNone
DeerFlow Smoke Testbytedance/deer-flow84k—~2.5kAutomated safety check: NotesMIT

Similar skills

  • Desktop Release Sync Landing

    Eynzof/Hermes-CN-Desktop

    Use only for stable/public Hermes Agent CN Desktop releases that should be visible to all users.

    1.8k GitHub stars~1.3k tokensUpdated 18 days ago
    DevOps & CloudAuto-check passed
  • Kouchou AI Development

    digitaldemocracy2030/kouchou-ai

    Local development setup, build and lint commands, environment configuration, and deployment helpers for the kouchou-ai repo.

    171 GitHub stars~540 tokensUpdated today
    DevelopmentAuto-check: notes
  • Deploy Release Test

    vercel/next.js

    Official

    Validate a commit-specific Next.js preview package and manually trigger the entire Next.js deployment test suite through the teste2edeployrelease.yml GitHub Actions workflow.

    143k GitHub stars~1.2k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Svelte Deployment

    spences10/svelte-claude-skills

    Svelte deployment guidance. An agent skill from spences10/svelte-claude-skills.

    218 GitHub stars~380 tokensUpdated today
    DevOps & CloudAuto-check passed
  • DeerFlow Smoke Test

    bytedance/deer-flow

    Walks through an end-to-end smoke test of a DeerFlow deployment: pull the latest code, deploy with Docker or locally, verify services, run health checks and write a report.

    84k GitHub stars~2.5k tokensUpdated today
    Testing & QAAuto-check: notes
  • TS SDK Author

    mindfold-ai/Trellis

    Design, build, verify, and publish production-grade TypeScript SDKs as npm packages inside a pnpm monorepo.

    15k GitHub stars~6.1k tokensUpdated 10 days ago
    DevelopmentAuto-check passed

More from different-ai/openwork

All 33 skills in this repo
  • OpenWork Desktop CDP Driver

    different-ai/openwork

    Drives a running OpenWork desktop window over CDP from the shell to evaluate JS, take screenshots, start sessions and send prompts for hand checks.

    24k GitHub stars~465 tokensUpdated today
    Auto-check passed
  • Fake Model Provider Faults

    different-ai/openwork

    Makes the desktop app's model provider fail on demand, with refused connections, resets, stalls and HTTP 4xx and 5xx errors, so error and retry states can be reproduced.

    24k GitHub stars~642 tokensUpdated today
    Auto-check passed
  • OpenWork Model Alias Manager

    different-ai/openwork

    Manages OpenWork's inference model aliases, discounts and overlays over the upstream OpenRouter catalog, and triggers the GitHub workflow that refreshes base models.

    24k GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Reproduce Chat States

    different-ai/openwork

    Fires known chat states in the running OpenWork desktop app, such as provider errors, retries and tool steps, so you can check how each renders.

    24k GitHub stars~673 tokensUpdated today
    Auto-check passed
  • Attaches OpenCode browser tools to the OpenWork Electron dev app through CDP to explore its UI, send a composer task and debug, not to give test verdicts.

    24k GitHub stars~780 tokensUpdated today
    Auto-check passed
  • Daytona Sandbox Operations

    different-ai/openwork

    Covers Daytona CLI setup, sandbox debugging, keeping a sandbox alive and which credentials the CLI uses, for when Daytona itself is the problem rather than the tests.

    24k GitHub stars~917 tokensUpdated today
    Auto-check passed

Works with

Questions about OpenWork PR Preview Manager

What does OpenWork PR Preview Manager do?

Boots, reopens, updates or resets disposable OpenWork preview environments for a pull request, locally or on a private Daytona or Freestyle host. This skill drives the repository's world lifecycle to give a pull request a running preview for hands-on testing. The available worlds are preview-desktop for the app only, preview-den for Den only, preview-full for Den plus the desktop app, and preview-app-web, and each can run locally, on a private Daytona URL or on Freestyle.

When should I use OpenWork PR Preview Manager?

OpenWork PR Preview Manager fits situations like: booting a preview of an OpenWork pull request for manual testing; reopening or updating an existing preview after new commits; choosing between desktop-only, Den-only and full previews; running a preview on a private Daytona URL or on Freestyle instead of locally.

How do I install OpenWork PR Preview Manager in Claude Code?

Run `npx skills add different-ai/openwork --skill preview-my-work -a claude-code`. Or copy the skill folder (.opencode/skills/preview-my-work in different-ai/openwork) into .claude/skills/preview-my-work in your project. Claude Code loads it when a task matches its description.

How do I install OpenWork PR Preview Manager in Codex?

Run `npx skills add different-ai/openwork --skill preview-my-work -a codex`. Or copy the skill folder (.opencode/skills/preview-my-work in different-ai/openwork) into .agents/skills/preview-my-work in your project. Codex loads it when a task matches its description.

Can I use OpenWork PR Preview Manager in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add different-ai/openwork --skill preview-my-work -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/preview-my-work, .gemini/skills/preview-my-work, .github/skills/preview-my-work and .opencode/skills/preview-my-work in your project.

What does OpenWork PR Preview Manager need to run?

Going by SKILL.md and its folder, OpenWork PR Preview Manager needs Python for the scripts in its folder, the command-line tools its instructions call (pnpm) and credentials named DAYTONA_API_KEY and FREESTYLE_API_KEY. Our summary lists: pnpm and a checkout of the OpenWork repository; A Daytona API URL and key, or a Freestyle API key, for remote previews; Infisical CLI access to fetch those keys, if you use the team setup.

Does OpenWork PR Preview Manager access the network?

SKILL.md names 1 domain. As links in the text: daytona.io. This is read from the text; nothing was executed.

Is OpenWork PR Preview Manager safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does OpenWork PR Preview Manager use?

OpenWork PR Preview Manager has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does OpenWork PR Preview Manager use?

About 5.9k tokens (SKILL.md is roughly 24k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to OpenWork PR Preview Manager?

Skills that share tags, products or a category with OpenWork PR Preview Manager: Desktop Release Sync Landing (Eynzof/Hermes-CN-Desktop, 1.8k stars), Kouchou AI Development (digitaldemocracy2030/kouchou-ai, 171 stars), Deploy Release Test (vercel/next.js, 143k stars) and Svelte Deployment (spences10/svelte-claude-skills, 218 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains OpenWork PR Preview Manager?

different-ai (a GitHub organization) maintains it in different-ai/openwork, which has 23,967 GitHub stars. The repository holds 33 skills in this directory. The repository was last updated on October 9, 2026.

Source: different-ai/openwork on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.