Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Vironnimo/vbot --skill computer-usegit clone --depth 1 https://github.com/Vironnimo/vbotWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vironnimo/vbot/computer-use)<a href="https://agentmods.dev/skills/vironnimo/vbot/computer-use"><img src="https://agentmods.dev/badge/skills/vironnimo/vbot/computer-use/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vironnimo/vbot/computer-use"><img src="https://agentmods.dev/badge/skills/vironnimo/vbot/computer-use.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.01278 |
| Opus 5.5 | $0.00014 | $0.00511 |
| Sonnet 5.5 | $0.00007 | $0.00256 |
| Haiku 4.5 | $0.00003 | $0.00128 |
Grade A, and why
computer-use scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Computer Use
computer moves the user's real mouse and keyboard on the desktop of the computer the vBot server runs on. It is slow and visible to the user, so use it only when nothing better fits:
- Prefer a shell, files, an app's command line or an API when they can do the job.
- For web pages, prefer the playwright-cli Skill if you have it; it is faster and more reliable than clicking through a browser.
Workflow
- If computer_apps is available,
{"action":"list"}shows displays, running apps and whether the user requires approval per app; add"query"to search installed apps. Otherwise take a screenshot of the whole monitor with computer{"action":"screenshot","view":"display"}. - Only when approval per app is required: computer_apps
{"action":"request","apps":["Notepad"],"reason":"..."}asks the user and waits for the answer. Name every app the task needs in one request. If computer_apps is unavailable, ask the user to enable it before operating an unapproved app. If the user declines, do not ask again unless they tell you to. - If computer_apps is available,
{"action":"open","app":"Notepad"}brings the app to the front, starting it if needed, and returns a screenshot. Otherwise click the app in the taskbar or Start menu of a display screenshot. - Act with computer, one action per call. If computer_batch is available, use it for predictable steps, such as click a field, type, press enter. The steps of a batch return no screenshots in between, so batch only steps that do not depend on what an earlier step shows. A batch that sent input ends with one screenshot; to check fine details, end the batch with a zoom of the area you worked on.
- Check the screenshot each result returns before the next step. A successful input call means the input was sent; verify that the intended control, color, text or drawing actually changed before building on it. When a call fails after it sent input, its result shows the screen as the input left it.
- When done, use computer_apps
{"action":"release"}if available. Otherwise finish your reply, which ends control.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago Changed · +1 lines 10904da47e19
- 13d ago Changed · -7 lines · +11 tokens per session 9726af30a5e4
- 20d ago Changed · +3 lines f720b05366f4
- 1mo ago Changed · +10 lines 8f7d1f50a981
- 1mo ago Changed · +8 lines 632c1cf90eeb
- 1mo ago Changed · -31 lines · -36 tokens per session 71fbd8c7ba73
- 1mo ago First seen · 66 lines · 59 tokens per session scan A c0dd26da9efa
computer-use is a skill published in the GitHub repository Vironnimo/vbot (1 stars, last pushed yesterday), licensed Apache-2.0. It adds 34 tokens to every session and 1,278 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
django-patterns
Django architecture patterns, REST API design with DRF, ORM best practices, caching, signals, middleware, and production-grade Django apps.
accelerate
Run PyTorch training across GPUs with minimal changes.
fastapi-patterns
FastAPI patterns for async APIs, dependency injection, Pydantic request and response models, OpenAPI docs, tests, security, and production readiness.
python-patterns
Python-specific design patterns and best practices including protocols, dataclasses, context managers, decorators, async/await, type hints, and package organization. Use when working with Python code to apply Pythonic patterns.
python-testing
Python testing best practices using pytest including fixtures, parametrization, mocking, coverage analysis, async testing, and test organization. Use when writing or improving Python tests.
pytorch-patterns
PyTorch deep learning patterns and best practices for building robust, efficient, and reproducible training pipelines, model architectures, and data loading.