Skills/npm/computer-use-linux

computer-use-linux@0.7.47

VERIFIED· install only

Publisher claims: "Native Linux desktop control for AI agents (MCP server, Pi tools, and CLI): semantic AT-SPI accessibility trees, screenshots, compositor window targeting, Wayland/X11 input, and an on-screen activity indicator." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher agent-sh · publishing for 184 daysInstalls 12,070Last detonated 7d ago
$claude mcp add computer-use-linux -- npx -y @agent-sh/computer-use-linux
WatchReport an issue

Coverage

Nothing executable in the bundle ran; the trace covers the install step only.
Tested
second launch with the clock 45 days ahead · every file read, file written and host contacted by what ran
Not tested
use of the skill by an agent

What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

Code inventory arrives with this version's next honeypot run.
It will ask for
unknown until the next run
Where it talks to
unknown until the next run
What the honeypot saw
no credential files read · no network
What we could not test
use of the skill by an agent

Not enough was observed to say yet. Nothing executable in the bundle ran; the trace covers the install step only. This version's code has not been inventoried for credentials and hosts either; both arrive with its next honeypot run (popular servers first). `vet-skills try computer-use-linux` runs it now and prints the full answer.

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-07 and scored it 92/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 3 s · 2 scenarios · 0 deviations
t+CategoryObserved action
1.7sprocessscript backend runs setup only
1.7sprocessscript backend runs setup only
2.1sprocesssecond launch, clock +45 days, CI unset
2.6sprocessexec /usr/bin/faketime

Static findings

MEDIUM
File type combined with upload actionpi/extension/mcp-client.bundle.cjs:64 · File type combined with upload action — agent may exfiltrate local data files|| ${N} === "boolean" || ${S} === null`).assign(F,(0,n._)`[${S}]`)}}}function b({gen:y,parentData:w,parentDataProperty:R},v){y.if((0,n._)`${w} !== undefined`,()=>y.assign((0,n._)`${w}[${R}]`,v))}functFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07

No other findings across 5,976 lines in 14 files.

Cost to the agent

Instructions loaded
≈5,301 tok
1 file · 14 in bundle

Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.

Known vulnerabilities

No advisories in OSV for @agent-sh/computer-use-linux@0.7.47 as of 7d ago. This is the package itself, not its dependencies.