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 jeffh/claude-plugins --skill computergit clone --depth 1 https://github.com/jeffh/claude-pluginsWrote 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/jeffh/claude-plugins/computer)<a href="https://agentmods.dev/skills/jeffh/claude-plugins/computer"><img src="https://agentmods.dev/badge/skills/jeffh/claude-plugins/computer/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/jeffh/claude-plugins/computer"><img src="https://agentmods.dev/badge/skills/jeffh/claude-plugins/computer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
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 →
- high Excessive Agency · line 44 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
- medium Excessive Agency · line 3 Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.Fix: Remove the model/provider override or disclose it prominently and require explicit operator approval before invoking an external coding CLI or billed model.
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 | $0.00168 | $0.01675 |
| Opus 5.5 | $0.00067 | $0.00670 |
| Sonnet 5.5 | $0.00034 | $0.00335 |
| Haiku 4.5 | $0.00017 | $0.00168 |
Grade A, and why
computer 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 8d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Computer Use
Delegate a desktop UI task to a Codex subagent running GPT 6.1 Sol (or another GPT model the user names) with Codex's Computer Use feature. The subagent will take screenshots and drive the Mac by clicking, typing, and scrolling.
Choosing the model
The -m flag selects the model. Default to gpt-6.1-sol, but honor any specific model the user asks for:
- Use
gpt-6.1-solunless the user names a different model. - If the user specifies a model — e.g. "drive it with gpt-5.6-terra", "use the
<name>model" — pass that exact string to-minstead. Don't validate or second-guess the name; Codex will error if it's unknown. Note the model must support Computer Use; if it doesn't, the run will fail and you should report that back. - If they typed
/codex:computer --model <name> <task>(or-m <name>), strip that flag from the prompt and use<name>as the model.
Choosing the effort
Reasoning effort is set with -c model_reasoning_effort="<level>". Default to low for Computer Use — each action is a screenshot + reasoning cycle, and low effort keeps the loop fast. Honor any level the user asks for:
- Use
lowunless the user names a different level. - If the user asks in prose — e.g. "high effort" — substitute that level.
- If they typed
/codex:computer --effort <level> <task>, strip that flag from the prompt and use<level>as the effort.
The command below shows -m gpt-6.1-sol and low effort; substitute the chosen model and effort.
Before invoking — confirm intent
Computer Use has real-world side effects (sending messages, making purchases, changing settings). Before spawning the subagent:
- Confirm the task is genuinely a GUI task, not something better done via CLI or code.
- If the task could touch sensitive scopes — sending messages, transmitting personal data, deleting cloud data, submitting forms, entering passwords, installing software, changing system settings — surface that risk to the user and get explicit approval before invoking. Because the run bypasses approvals and the sandbox (see below), Codex will not stop to confirm risky steps mid-run — Claude's up-front confirmation is the only gate, so build any "don't do X without confirming" constraint into the prompt itself.
- The Codex Computer Use app must be present at
~/.codex/computer-use/Codex Computer Use.app. If it's missing, tell the user to install it via the Codex desktop app (codex app).
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.
- 8d ago Changed 795ca466c83a
- 1mo ago Changed · +3 lines 995b2b6f513d
- 1mo ago First seen · 76 lines · 168 tokens per session scan A d03be2e308e9
computer is a skill published in the GitHub repository jeffh/claude-plugins (12 stars, last pushed 9d ago), licensed Apache-2.0. It adds 168 tokens to every session and 1,675 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
interactive-login
Log a CLI in with a browser/device-code flow (aws sso, gh, glab, gcloud, …) and save the result to the keychain so later commands can request it through execute.credentials.
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.
azure-messaging-webpubsub-java
Build real-time web applications with Azure Web PubSub SDK for Java. Use when implementing WebSocket-based messaging, live updates, chat applications, or server-to-client push notifications.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
open-design-browser
Inspect and exercise local HTML design artifacts with the existing agent-browser CLI, including screenshots and keyboard or interaction checks. Use when rendered evidence helps author or review a design; excludes browser installation, remote services, and unrelated browser automation.