Which versions of anthropic-setup are malicious?

The following npm versions of anthropic-setup are flagged malicious: 1.0.0, 1.0.1. Treat any of these as compromised.

How do I remediate anthropic-setup if I installed it?

anthropic-setup is a typosquat — you almost certainly intended a legitimately-named package. Remove anthropic-setup, install the correct package, and rotate any secrets exposed during the install since post-install scripts may have already run.

I already installed anthropic-setup — how do I know if it already compromised me?

If anthropic-setup was installed, its post-install payload may already have run. Removing the package does not undo that — check outbound connections and credential use from the install window onward.

Is anthropic-setup part of a known attack campaign?

Yes — anthropic-setup is associated with the IN-MAL-2026-014938, GHSA-6mwf-mqp4-2f69, IN-MAL-2026-017992 campaign. Packages in the same campaign often share infrastructure and payloads, so check whether any related packages are also in your dependency tree.

How do I find anthropic-setup across my projects, lockfiles, and CI pipelines?

Search your lockfiles and build artifacts for anthropic-setup (2 malicious versions).

Malicious package

anthropic-setupnpm

Advisory published Aug 5, 2026Updated Aug 18, 2026

anthropic-setup is a confirmed malicious npm package (MAL-2026-12510) that typosquats a legitimate package to trick installs (malicious versions 1.0.0, 1.0.1). Do not install it — remove it immediately and rotate any exposed credentials.

Malicious code in anthropic-setup (npm)

MAL-2026-12510

Immediate action

Remove the package, then rotate any secrets the build/runtime could reach.

npm uninstall anthropic-setup

What this malware does

The package is presented as an Anthropic API setup helper (npx anthropic-setup sk-ant-...). When invoked, it writes ~/.claude/settings.json with ANTHROPIC_BASE_URL set to https://sugarball.vercel.app, stores the user-supplied sk-ant- API key in ANTHROPIC_API_KEY, and configures an apiKeyHelper of echo '<key>'. After this configuration, the official Claude CLI sends every prompt and the user's Anthropic API key to sugarball.vercel.app instead of api.anthropic.com. The package name and description impersonate official Anthropic tooling, reinforcing the deception. The redirect destination is hardcoded and is not the installer's own infrastructure, and this behavior is not disclosed as the package's purpose — the user provides their key expecting official Anthropic use.

Any computer that has this package installed or running should be considered fully compromised. All secrets and keys stored on that computer should be rotated immediately from a different computer. The package should be removed, but as full control of the computer may have been given to an outside entity, there is no guarantee that removing the package will remove all malicious software resulting from installing it.

Malicious versions

2 flagged

1.0.01.0.1

Indicators of compromise (SHA-256)

9a5019e390fffd163e88f49fe1fc12c4ce8d2e50d7563cf7cf0a38508dc4eef5

229b8ab0ffc87ae8a12a3f55622a684e0765eeef6e84f1ed85ac48ab96080c54

baee404f50ce2f767660b7f6cf21dec0df6b0d0da68b9bc39ce506f941f1f664

Detection & response playbook

Typosquat

Find it
Search your lockfiles and build artifacts for anthropic-setup (2 malicious versions).
If you installed it — respond
anthropic-setup is a typosquat — you almost certainly intended a legitimately-named package. Remove anthropic-setup, install the correct package, and rotate any secrets exposed during the install since post-install scripts may have already run.
Did it already run?
If anthropic-setup was installed, its post-install payload may already have run. Removing the package does not undo that — check outbound connections and credential use from the install window onward.

Frequently asked questions

No. anthropic-setup on npm has been identified as a malicious package (versions 1.0.0, 1.0.1 flagged). It should be removed immediately — do not install or keep it in your dependency tree.

Campaign

IN-MAL-2026-014938GHSA-6mwf-mqp4-2f69IN-MAL-2026-017992

References

Credits

Amazon Inspector · finder

Detect & block this

O3 blocks anthropic-setup-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the malicious outbound activity and severs the channel.

Malware detection Runtime egress monitoring

Explore

All malware advisories How O3 detects malware Vulnerability database