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 yutori-ai/yutori-mcp --skill 06-computer-usegit clone --depth 1 https://github.com/yutori-ai/yutori-mcpWrote 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/yutori-ai/yutori-mcp/06-computer-use)<a href="https://agentmods.dev/skills/yutori-ai/yutori-mcp/06-computer-use"><img src="https://agentmods.dev/badge/skills/yutori-ai/yutori-mcp/06-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/yutori-ai/yutori-mcp/06-computer-use"><img src="https://agentmods.dev/badge/skills/yutori-ai/yutori-mcp/06-computer-use.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 9 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 17 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 18 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 22 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 27 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 45 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 60 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 67 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 112 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
- medium MCP Rug Pull · line 118 uvx/uv tool run commands without ==version create a rug-pull risk.Fix: Pin the version: uvx package-name==1.2.3
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 | $0.00060 | $0.01289 |
| Opus 5 | $0.00030 | $0.00645 |
| Sonnet 5 | $0.00012 | $0.00258 |
| Haiku 4.5 | $0.00006 | $0.00129 |
Grade A, and why
yutori-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 today.
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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Computer Use
Run tasks on the user's visible Mac desktop with Yutori computer use, or drive a single app's window in the background while the user keeps working.
Setup
If the user has not set up computer use yet, ask them to run these commands in a real terminal:
uvx yutori-mcp login
uvx yutori-mcp computer-use setup
setup ends with the readiness checks and reports anything still missing. To re-check later,
or when a run fails, use uvx yutori-mcp computer-use doctor.
Use the smoke test when validating a fresh install:
uvx yutori-mcp computer-use smoke
Requirements: macOS 15+, Python 3.10+, uvx, a Yutori API key with computer-use access,
CuaDriver.app / cua-driver==0.28.0, and Screen Recording plus Accessibility permissions.
The optional native reasoning overlay may require Xcode Command Line Tools.
Run a Task
In Claude Code, including Claude sessions hosted by Conductor, run the CLI through the Bash
tool with stdout attached. This makes runner ready, each action #... line, shell-command
previews, and the final result appear in the agent's live progress. Do not redirect the command
to a file or call the MCP run_computer_use_task tool from these sessions: Claude's headless
Agent SDK does not expose MCP progress or log notifications to its host.
For a foreground run:
uvx yutori-mcp computer-use run "$ARGUMENTS" --minutes 9 --max-steps 60
Run that as a foreground Bash call and set the Bash tool timeout slightly longer than the CLI's
--minutes limit (Claude's foreground Bash timeout is capped at 10 minutes). Pass the task as
one safely quoted argument and do not use eval.
For a longer run, start the same CLI command as a background Bash task without redirecting its
output, then wait on it with TaskOutput in blocking intervals. Relay meaningful new action
lines in progress updates between waits. The computer-use --minutes value remains the absolute
run deadline.
For an app-specific task:
uvx yutori-mcp computer-use run "$ARGUMENTS" --app Safari --start-url https://example.com --minutes 9
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.
- today Changed 2e04a16706b2
- 7d ago Changed · +23 lines 32415df67968
- 8d ago Changed · +32 lines · +15 tokens per session 94a7156b38e3
- 12d ago First seen · 77 lines · 45 tokens per session scan A ec93c89c1bd6
yutori-computer-use is a skill published in the GitHub repository yutori-ai/yutori-mcp (24 stars, last pushed today), licensed Apache-2.0. It adds 60 tokens to every session and 1,289 once invoked, about $0.0003 per session on Opus 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 2026-08-30.
Other skills, from other repositories
site-audit
Use when user asks to audit a website, review all pages, check for broken links, capture full site state, or do a comprehensive product review.
kaboom
Use when user asks to check browser state, debug page errors, inspect network traffic, take screenshots, automate clicks or form fills, run accessibility or security audits, measure performance, generate tests, or record browser sessions. Trigger phrases: "check my browser", "debug this page", "take a screenshot"…
automate
Use when user asks to automate browser actions, click buttons, fill forms, navigate pages, run a sequence of browser steps, or build a repeatable workflow.
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.
agent-user-tools
Agent skill for user-tools - invoke with $agent-user-tools.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.