Which versions of anthropic-internal-tools are malicious?

The following npm versions of anthropic-internal-tools are flagged malicious: 1.0.0, 1.0.1, 1.0.2. Treat any of these as compromised.

How do I remediate anthropic-internal-tools if I installed it?

anthropic-internal-tools is built to steal secrets, so assume every credential the build or runtime could read is compromised. Remove it from your project and lockfile, then rotate ALL exposed secrets — npm/registry tokens, cloud keys, CI/CD secrets, SSH keys, and any .env values — from a known-clean machine. Audit logs for unauthorized use of those credentials.

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

If anthropic-internal-tools was ever installed, its post-install/runtime payload may have already executed. O3's L7 egress monitoring and runtime eBPF sensors detect the credential exfiltration or command-and-control callback after install and block the malicious outbound channel, so you catch and contain the actual compromise — not just the presence of the package.

Is anthropic-internal-tools part of a known attack campaign?

Yes — anthropic-internal-tools is associated with the IN-MAL-2026-007688, IN-MAL-2026-007690, IN-MAL-2026-007689 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-internal-tools across my projects, lockfiles, and CI pipelines?

Scan your lockfiles (package-lock.json, pnpm-lock.yaml, yarn.lock, requirements.txt, poetry.lock, etc.) and build artifacts for anthropic-internal-tools (3 malicious versions). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging anthropic-internal-tools across your stack and pipelines.

Malicious package

anthropic-internal-toolsnpm

Malicious code in anthropic-internal-tools (npm) Remove it immediately and rotate any exposed credentials.

MAL-2026-6551

Immediate action

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

npm uninstall anthropic-internal-tools

What this malware does

Package name impersonates the Anthropic namespace and ships a preinstall hook (scripts.preinstall = 'node index.js') that executes on every npm install. index.js performs bulk reads of installer-side credential files from the home directory — ~/.aws/credentials, ~/.aws/config, ~/.config/gcloud/application_default_credentials.json, ~/.azure/accessTokens.json, ~/.ssh/id_rsa.pub (and probes id_rsa), ~/.npmrc, ~/.gitconfig — and uses execSync to curl AWS IMDS (http://169.254.169.254/latest/meta-data/iam/security-credentials/) and GCP metadata (http://metadata.google.internal, Metadata-Flavor: Google) to capture IAM/service-account tokens. It also collects os.hostname(), os.userInfo(), cwd, and environment variables matching KEY|TOKEN|SECRET|PASS|AUTH|CRED|AWS|GCP|AZURE|NPM|REGISTRY. The beacon is POSTed via https.request to a hardcoded collector at https://webhook.site/2d1764b2-1249-4793-840f-7846d7d820cd. Installing this package on a developer workstation or CI runner discloses long-lived cloud credentials, SSH keys, and registry tokens to a third-party endpoint, and on cloud-hosted CI additionally yields short-lived IAM/service-account tokens usable to pivot into the installer's cloud account. The package self-describes as a 'dependency confusion PoC', confirming the namespace-impersonation intent against an internal Anthropic-named package.

Malicious versions

3 flagged

1.0.01.0.11.0.2

Indicators of compromise (SHA-256)

47963d44c126223c729c2c53a19c1c7d79f2f66cc1dca56c98ba4412eed31f5f

63383c78f8dd92a67e70a3599f18778ff6ce73e42cba64abec624522ea1a254b

ab3bb04aee6f5f1d8768b7fd2173cd7c0cac18b5d83d6a83cf2be96a7512d8f7

Detection & response playbook

Credential / info stealer

Find it
Scan your lockfiles (package-lock.json, pnpm-lock.yaml, yarn.lock, requirements.txt, poetry.lock, etc.) and build artifacts for anthropic-internal-tools (3 malicious versions). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging anthropic-internal-tools across your stack and pipelines.
If you installed it — respond
anthropic-internal-tools is built to steal secrets, so assume every credential the build or runtime could read is compromised. Remove it from your project and lockfile, then rotate ALL exposed secrets — npm/registry tokens, cloud keys, CI/CD secrets, SSH keys, and any .env values — from a known-clean machine. Audit logs for unauthorized use of those credentials.
Did it already run?
If anthropic-internal-tools was ever installed, its post-install/runtime payload may have already executed. O3's L7 egress monitoring and runtime eBPF sensors detect the credential exfiltration or command-and-control callback after install and block the malicious outbound channel, so you catch and contain the actual compromise — not just the presence of the package.
How O3 protects you
O3 blocks anthropic-internal-tools before install through its supply-chain scanner, and if it has already run, detects and severs the exfiltration or C2 callback at runtime through L7 egress monitoring and eBPF.

Frequently asked questions

No. anthropic-internal-tools on npm has been identified as a malicious package (versions 1.0.0, 1.0.1, 1.0.2 flagged). It should be removed immediately — do not install or keep it in your dependency tree.

Campaign

IN-MAL-2026-007688IN-MAL-2026-007690IN-MAL-2026-007689

References

Credits

Amazon Inspector · finder

Detect & block this

O3 blocks anthropic-internal-tools-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the credential exfiltration and severs the channel.

Malware detection Runtime egress monitoring