New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Setup Osworld

by xlang-ai in xlang-ai/OSWorld-V2

Provision and verify an OSWorld-V2 checkout after clone. An agent skill from xlang-ai/OSWorld-V2.

Apache-2.0Auto-check: notesDevOps & Cloud

Install Setup Osworld

skills CLI
$ npx skills add xlang-ai/OSWorld-V2 --skill setup-osworld -a claude-code

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

GitHub CLI
$ gh skill install xlang-ai/OSWorld-V2 setup-osworld --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/xlang-ai/OSWorld-V2.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.codex/skills/setup-osworld .claude/skills/setup-osworld && 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
setup-osworld
GitHub stars
355
Used in
1 other repo
Token cost
~2.6k tokens
SKILL.md length
1,235 words
Files
12 (incl. references)
Skills in repo
4
Repo updated
First seen
Licence
Apache-2.0

At a glance

Provision and verify an OSWorld-V2 checkout after clone. An agent skill from xlang-ai/OSWorld-V2.

  • Works in 8 steps: Checkout and documentation. → Dependencies. Read references/common.md… → Provider. Load exactly one of… → …
  • The user asks for OSWorld-V2 setup
  • SKILL.md covers Required Version Argument, Supported Scope, Start With Intake and Versioned References, plus 2 more sections
  • Calls uv; needs GITLAB_PRIVATE_TOKEN

What it does

Setup Osworld is an agent skill from xlang-ai/OSWorld-V2. Provision and verify an OSWorld-V2 checkout after clone. Use when the user asks for OSWorld-V2 setup, installation, onboarding, AWS provider setup, Docker provider setup, mocked website server setup, GitLab server setup, gated task download, or a final runnable export block. The skill should install/configure the selected supported infrastructure where possible, ask for user confirmation or credentials when required, and report what is fully configured versus still blocked.

Its SKILL.md is about 2.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 13 other files, including reference files (for example `agents/openai.yaml`, `references/common.md` and `references/gitlab.md`).

It sits in DevOps & Cloud, covering Linux administration and Containers. It works with Amazon Web Services, Docker and GitLab. The repository describes itself as: OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks. The licence is Apache-2.0.

When your agent uses it

  • The user asks for OSWorld-V2 setup
  • AWS provider setup
  • Docker provider setup
  • Mocked website server setup

Example prompts

  • “/setup-osworld”

Requirements

  • Python 3
  • Docker
  • A credential in GITLAB_PRIVATE_TOKEN

Workflow steps

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

  1. Checkout and documentation.
  2. Dependencies. Read references/common.md for tool checks.
  3. Provider. Load exactly one of references/provider-aws.md,
  4. Websites.
  5. GitLab and proxy. Read references/gitlab.md and/or references/proxy.md
  6. Tasks and assets.
  7. Verification. Load references/verify.md. For either version, verify code,
  8. Final outputs. Report the selected --version, exact benchmark release,

What it can do on your machine

Read from SKILL.md and the folder at commit acdd349. 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

    Shell commands in SKILL.md call:

    • uv

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

  • Network

    No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.

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

  • Credentials

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

    • GITLAB_PRIVATE_TOKEN

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

Context cost

Setup Osworld loads about 2.6k tokens when it runs, and up to ~12k if it reads all its reference files. Until then it costs about 123 tokens; SKILL.md has 1,235 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~123
When it runs · the whole SKILL.md, loaded when a task matches
~2.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~12k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:179
    - Never commit secrets, `.env` files, generated private keys, or setup export

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); files beside SKILL.md are not scanned.

SKILL.md

The full file from xlang-ai/OSWorld-V2 at commit acdd349, republished under its Apache-2.0 licence (© xlang-ai). 1,235 words, ~2,599 tokens.

Download SKILL.mdSave it as .claude/skills/setup-osworld/SKILL.md (or your agent's skills folder). This skill also uses 11 other files; get the full folder from GitHub.
name
setup-osworld
description
Provision and verify an OSWorld-V2 checkout after clone. Use when the user asks for OSWorld-V2 setup, installation, onboarding, AWS provider setup, Docker provider setup, mocked website server setup, GitLab server setup, gated task download, or a final runnable export block. The skill should install/configure the selected supported infrastructure where possible, ask for user confirmation or credentials when required, and report what is fully configured versus still blocked.

Setup OSWorld

Use this skill to make a cloned OSWorld-V2 checkout runnable, not merely to list environment variables. Provision every selected component that can be configured safely from the current machine, and ask the user whenever an action requires credentials, paid resources, DNS changes, SSH keys, secrets, or a destructive/cloud operation.

Required Version Argument

Require --version 2 or --version 2.1 in the skill invocation before any setup commands, package installation, downloads, checkout changes or provisioning. There is no default. If missing, ask the user to supply the argument and wait. Reject unsupported values or conflicting duplicate values; ask for correction. An explicitly supplied argument earlier in the same setup session remains valid. Do not infer it from the current checkout, README, or a bare benchmark tag.

These are arguments to this skill's instruction workflow, not a shell command or new flags for the Python download scripts.

text
$setup-osworld --version 2.1
$setup-osworld --version 2 --benchmark-release osworld-v2-2026.08.08
$setup-osworld --version 2 --benchmark-release osworld-v2-2026.06.24
ArgumentRelease selectionBehavior
--version 2.1osworld-v2.1Use the osworld-v2.1 manifest and references. An optional --benchmark-release must equal osworld-v2.1.
--version 2osworld-v2-2026.08.08 or osworld-v2-2026.06.24Require --benchmark-release; ask and wait if omitted. Use that historical manifest and matching checkout.

Resolve the selected manifest before proceeding. Reject a version/release mismatch, main, latest, or an unsupported release rather than silently falling back. Preserve user changes when checking out the selected code; use a separate checkout when necessary. Keep the skill and its references accessible from the original checkout; resolve runtime code and manifests in the target checkout. Carry the resolved version and release through all later steps, including final exports and verification.

Supported Scope

Supported provider setup:

  • aws: fully provision required AWS network resources with AWS CLI, then export the values OSWorld needs.
  • docker: install/verify Docker and KVM where possible, then verify the Docker provider can run.

Unsupported provider setup for now:

  • vmware, virtualbox, azure, gcp, aliyun, and volcengine.

For unsupported providers, load references/unsupported-providers.md, state that this skill does not automate them yet, and ask whether the user wants AWS or Docker instead.

Optional service setup:

  • OSWorld-web mocked website server, using Task-Web/OSWorld-web.
  • GitLab server, using Task-Web/gitlab.
  • Gated task class download from Hugging Face.
  • Version-pinned task asset download from gated Hugging Face storage.
  • Task-scoped proxy setup following upstream OSWorld section 2.3, plus optional host-side proxy or HF mirror setup.

Start With Intake

First satisfy the required version argument above. Then ask before doing infrastructure work unless the user has already supplied the answers.

Ask for:

  • provider: aws or docker
  • whether to use existing AWS resources or create new ones
  • AWS region, VPC/subnet preference, and whether AWS charges are acceptable
  • whether to set up OSWorld-web, and on existing server or new AWS EC2 host
  • website domain mode: existing wildcard domain, nip.io, or user-provided host suffix
  • whether to set up GitLab, and whether it must be a separate server/domain
  • whether to download gated tasks and complete assets for the selected release
  • whether task-scoped proxy, host-side proxy, or HF mirror is needed
  • whether to run real smoke tests that create/start resources

If the user says "do it unattended", still stop for any cloud spend, DNS, GitHub private-repo authorization, generated secret disclosure, or destructive operation.

Versioned References

Use only the column matching the required --version argument.

Referenceosworld-v2.1Historical 2.0 releases
Migration from OSWorld 1.0docs/MIGRATING_v2.1_FROM_OSWORLD_V1.mddocs/MIGRATING_FROM_OSWORLD_V1.md
Public evaluationdocs/PUBLIC_EVALUATION_GUIDELINE_v2.1.mddocs/PUBLIC_EVALUATION_GUIDELINE.md
Tasks and assetsevaluation_examples/task_class/README_v2.1.mdreferences/tasks.md
Websitesreferences/website_v2.1.mdreferences/website.md

Read the selected release manifest for exact component references. Historical examples may show a different date or an obsolete hosted suffix: use the selected historical manifest's tags and verify an existing deployment before adopting it. Do not edit the preserved historical reference files while executing setup.

Show full SKILL.md (642 more words)Show less

Workflow by Version

  1. Checkout and documentation.
    • 2.1: use code tag osworld-v2.1 and the osworld-v2.1 evaluation/migration guides.
    • 2: use osworld_code.tag from the selected date manifest and the historical evaluation/migration guides. Do not run from an osworld-v2.1 checkout.
    • In both cases, use docs/PROVIDER_SETUP.md only for shared prerequisites; the selected manifest controls release-specific values.
  2. Dependencies. Read references/common.md for tool checks.
    • 2.1: run uv sync --frozen; use uv sync --frozen --extra full only when the optional stack is needed.
    • 2: install from the selected historical checkout's lockfile with uv sync --frozen (and --extra full if requested). If that checkout cannot support the command, report the actual failure; do not borrow the osworld-v2.1 lockfile or silently update dependencies.
  3. Provider. Load exactly one of references/provider-aws.md, references/provider-docker.md, or references/unsupported-providers.md.
    • 2.1: use the osworld-v2.1 manifest's AMI or Docker artifact and runtime references.
    • 2: use the selected historical manifest's provider image references.
    • Check actual provider configuration against the selected manifest before launch. A common reference's example image must not override the manifest.
  4. Websites.
    • 2.1: load references/website_v2.1.md; deploy website tag osworld-v2.1 with its pinned submodules, or verify an existing deployment matches it.
    • 2: load references/website.md; use website_code.tag from the selected historical manifest, including its pinned submodules. Do not reuse an old hosted suffix solely because it appears in an example.
  5. GitLab and proxy. Read references/gitlab.md and/or references/proxy.md only when requested. These service procedures are shared; configure them for the tasks and website host of the selected release.
  6. Tasks and assets.
    • 2.1: load evaluation_examples/task_class/README_v2.1.md; both downloaders receive --benchmark-release osworld-v2.1. Use the full gated asset snapshot; the public asset repository is not its replacement.
    • 2: load references/tasks.md; available downloaders receive the explicitly selected date release, replacing generic or differently dated examples. v2026.06.24 does not contain download_osworld_v2_assets.py. For that checkout, use uv run python with huggingface_hub.snapshot_download: read assets.repository, assets.repo_type and assets.tag from the selected historical manifest and pass them as repo_id, repo_type and revision, with local_dir set to a fresh complete asset directory. Require gated access, download the full snapshot without allow/ignore filters, and export its absolute path as OSWORLD_FILE_BASE_URL. Do not copy a newer downloader into the historical checkout or fall back to main.
    • Never download floating main content or mix task and asset releases.
  7. Verification. Load references/verify.md. For either version, verify code, website, tasks, assets and provider references against its manifest, check the expected task count, and run only the authorized smoke tests. Use the task hash source specified by that manifest: osworld-v2.1 keeps the regenerated list in this code repository; historical releases retain their recorded sources.
  8. Final outputs. Report the selected --version, exact benchmark release, component refs and actual verification results. Export OSWORLD_BENCHMARK_RELEASE as osworld-v2.1 or the selected date release, respectively, alongside OSWORLD_FILE_BASE_URL and the chosen website suffix. Put secret exports in a chmod 600 local file. Report configured, skipped and blocked components separately; never imply that a different version was tested.

Operating Rules

  • Use uv, uvx, and uv run python; do not rely on bare python or pip.
  • Prefer existing repo docs and current provider code over memory.
  • Do not stop at "set this variable"; if the user selected a supported setup, create or verify the backing resource that makes the variable valid.
  • Use AWS CLI for AWS resources. Do not hand-wave AWS console steps unless the user chose manual setup.
  • Use Docker Compose for OSWorld-web and GitLab exactly as their READMEs describe, with HOST_SUFFIX, GITLAB_URL, and GITLAB_PRIVATE_TOKEN backed by reachable services.
  • Keep model/API-key setup separate from provider/runtime setup unless the user asks to run an agent or evaluator.
  • Never commit secrets, .env files, generated private keys, or setup export files.
  • If an external private repository is inaccessible, report the exact blocker and ask the user to authorize GitHub access instead of inventing commands.
  • At the end, be explicit: "configured and verified", "configured but not smoke-tested", "not configured", or "blocked awaiting user action".

© xlang-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 11 other files (references) in .codex/skills/setup-osworld of xlang-ai/OSWorld-V2.

  • SKILL.md
  • agents/openai.yaml
  • references/common.md
  • references/gitlab.md
  • references/provider-aws.md
  • references/provider-docker.md
  • references/proxy.md
  • references/tasks.md
  • references/unsupported-providers.md
  • references/verify.md
  • references/website.md
  • references/website_v2.1.md

Open the folder on GitHubat commit acdd349

Used in 1 other repository

We found 3 copies of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in xlang-ai/OSWorld-V2, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Setup Osworld 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.

Setup Osworld compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Setup Osworld this skillxlang-ai/OSWorld-V23551 repos~2.6kAutomated safety check: NotesApache-2.0
Senior DevOps Toolkitmaslennikov-ig/claude-code-orchestrator-kit2606 repos~1.1kAutomated safety check: NotesCustom licence
Minimegasandia-minimega/minimega160—~3.2kAutomated safety check: PassGPL-3.0-only
Setup Cpu Proxy Serverdrawthingsai/draw-things-community582—~3.8kAutomated safety check: PassGPL-3.0
Logfire Infrastructurepydantic/skills140—~1.8kAutomated safety check: PassMIT
Ksaildevantler-tech/ksail165—~1.1kAutomated safety check: PassCustom licence

Similar skills

  • Senior DevOps Toolkit

    maslennikov-ig/claude-code-orchestrator-kit

    Comprehensive DevOps skill for CI/CD, infrastructure automation, containerization, and cloud platforms (AWS, GCP, Azure). Includes pipeline setup…

    260 GitHub starsUsed in 6 repos~1.1k tokens
    DevOps & CloudAuto-check: notes

  • Minimega

    sandia-minimega/minimega

    This skill should be used when the user asks how to configure, run, automate, integrate, or troubleshoot minimega (VMs, namespaces, VLANs, clusters, miniccc, miniweb, command socket or Python API…

    160 GitHub stars~3.2k tokensUpdated 2 days ago
    DevOps & CloudAuto-check passed

  • Setup Cpu Proxy Server

    drawthingsai/draw-things-community

    Set up and verify a new Draw Things CPU proxy and Envoy server using the scripts in Scripts/ServerManagement/CPUScript.

    582 GitHub stars~3.8k tokensUpdated today
    DevOps & CloudAuto-check passed

  • Logfire Infrastructure

    pydantic/skills

    Official

    Monitor hosts, Docker containers, Kubernetes clusters, database/queue/cache servers, and cloud-provider metrics with Pydantic Logfire — no application code required.

    140 GitHub stars~1.8k tokensUpdated 8 days ago
    DevOps & CloudAuto-check passed

  • Ksail

    devantler-tech/ksail

    Use the ksail CLI to spin up and manage Kubernetes clusters (Kind/K3d/Talos/vCluster/KWOK — local via Docker; EKS — cloud via AWS) and GitOps workloads declaratively.

    165 GitHub stars~1.1k tokensUpdated today
    DevOps & CloudAuto-check passed

  • Agenticx Deployer

    DemonDamon/AgenticX

    Guide for deploying AgenticX agents to production including Docker containerization, Kubernetes orchestration, Volcengine AgentKit cloud deployment, and API server setup.

    315 GitHub stars~866 tokensUpdated today
    DevOps & CloudAuto-check passed

More from xlang-ai/OSWorld-V2

  • Migrate Osworld Agent

    xlang-ai/OSWorld-V2

    Migrate an agent from upstream OSWorld into this OSWorld-V2 repository, add matching evaluation entrypoints, and verify the integration.

    355 GitHub starsUsed in 1 repo~546 tokens
    Auto-check passed

  • Analyze Traj

    xlang-ai/OSWorld-V2

    Analyze OSWorld-V2 agent trajectory logs and task results to produce actionable insights.

    355 GitHub starsUsed in 1 repo~298 tokens
    Auto-check passed

  • Analyze Task

    xlang-ai/OSWorld-V2

    Check OSWorld tasks. An agent skill from xlang-ai/OSWorld-V2.

    355 GitHub starsUsed in 1 repo~223 tokens
    Auto-check passed

Works with

Categories

Questions about Setup Osworld

What does Setup Osworld do?

Provision and verify an OSWorld-V2 checkout after clone. An agent skill from xlang-ai/OSWorld-V2. Setup Osworld is an agent skill from xlang-ai/OSWorld-V2. Provision and verify an OSWorld-V2 checkout after clone.

When should I use Setup Osworld?

Setup Osworld fits situations like: the user asks for OSWorld-V2 setup; AWS provider setup; Docker provider setup; mocked website server setup.

How do I install Setup Osworld in Claude Code?

Run `npx skills add xlang-ai/OSWorld-V2 --skill setup-osworld -a claude-code`. Or copy the skill folder (.codex/skills/setup-osworld in xlang-ai/OSWorld-V2) into .claude/skills/setup-osworld in your project. Claude Code loads it when a task matches its description.

How do I install Setup Osworld in Codex?

Run `npx skills add xlang-ai/OSWorld-V2 --skill setup-osworld -a codex`. Or copy the skill folder (.codex/skills/setup-osworld in xlang-ai/OSWorld-V2) into .agents/skills/setup-osworld in your project. Codex loads it when a task matches its description.

Can I use Setup Osworld 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 xlang-ai/OSWorld-V2 --skill setup-osworld -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/setup-osworld, .gemini/skills/setup-osworld, .github/skills/setup-osworld and .opencode/skills/setup-osworld in your project.

What does Setup Osworld need to run?

Going by SKILL.md and its folder, Setup Osworld needs the command-line tools its instructions call (uv) and credentials named GITLAB_PRIVATE_TOKEN. Our summary lists: Python 3; Docker; A credential in GITLAB_PRIVATE_TOKEN.

Does Setup Osworld access the network?

SKILL.md contains no URLs. Its commands use uv, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Setup Osworld safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Setup Osworld use?

Setup Osworld is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Setup Osworld use?

About 2.6k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 9.9k tokens, read only when the agent opens those files.

What are the alternatives to Setup Osworld?

Skills that share tags, products or a category with Setup Osworld: Senior DevOps Toolkit (maslennikov-ig/claude-code-orchestrator-kit, 260 stars), Minimega (sandia-minimega/minimega, 160 stars), Setup Cpu Proxy Server (drawthingsai/draw-things-community, 582 stars) and Logfire Infrastructure (pydantic/skills, 140 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Setup Osworld?

xlang-ai (a GitHub organization) maintains it in xlang-ai/OSWorld-V2, which has 355 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 1, 2026.

Source: xlang-ai/OSWorld-V2 on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.