open-codex-computer-use is an open-source computer-use service exposed through MCP, allowing AI agents to control computers through accessibility features. It is for agents and MCP clients that need to perform actions on macOS, Linux, or Windows desktops. The catalogue includes instructions, a skill, and an MCP integration for using the service.
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.
git clone --depth 1 https://github.com/iFurySt/open-codex-computer-useInstalling this writes the configuration, not the program: `./scripts/launch-open-computer-use-repl.sh` has to be on your PATH already, or the server will not start.
Wrote 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/mcp/ifuryst/open-codex-computer-use/open-computer-use-open-computer-use)<a href="https://agentmods.dev/mcp/ifuryst/open-codex-computer-use/open-computer-use-open-computer-use"><img src="https://agentmods.dev/badge/mcp/ifuryst/open-codex-computer-use/open-computer-use-open-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/mcp/ifuryst/open-codex-computer-use/open-computer-use-open-computer-use"><img src="https://agentmods.dev/badge/mcp/ifuryst/open-codex-computer-use/open-computer-use-open-computer-use.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
open-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 26d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- open-computer-use — 100% identical, 2 lines differ
What it actually says
{
"open-computer-use": {
"command": "./scripts/launch-open-computer-use-repl.sh",
"cwd": "."
}
}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.
- 26d ago Changed bceb7782c52e
- 1mo ago First seen · 6 lines scan A 022d6687a01e
open-computer-use is an MCP server published in the GitHub repository iFurySt/open-codex-computer-use (2,489 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: an MCP server costs its tool schemas, not its config file. 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 mcp servers, from other repositories
axstream
Fast, verified computer use for agents: a streaming JSONL action spec, a verified click ladder, and macros that make repeat tasks instant. Runs locally from the axstream Python package.
agent-computer-use-mcp
A local MCP server for Gateway-managed computer use with cua-driver, OCR, and user-only desktop overlays. Runs locally from the agent-computer-use-mcp npm package.
mcp-server-browser
MCP server for browser use access. Runs locally from the @agent-infra/mcp-server-browser npm package. Needs 1 environment variable to run.
uisight
Your AI sees the screen but can't measure it — measured web UI findings, shared live sessions. Runs locally from the uisight npm package. Needs 3 environment variables to run.
uivision-mcp
MCP server for browser and desktop automation: OCR, image recognition, real mouse and keyboard input. Runs locally from the uivision-mcp-bridge npm package. Needs 1 environment variable to run.
hunch
Drive your Mac focus-free over MCP: OS APIs, AppleScript, CDP, and Accessibility. Runs locally from the hunch-sdk Python package.