@withgoogle/stitch-sdknpm
Malicious code in @withgoogle/stitch-sdk (npm) Remove it immediately and rotate any exposed credentials.
What this malware does
@withgoogle/stitch-sdk is a scope-squatting package on npm that impersonates Google's Stitch AI design tool SDK. The attacker registered the @withgoogle scope to mimic Google's withgoogle.com domain and published versions 0.1.1 and 0.1.2 under the account maximus-mcmillan on June 19, 2026. The package runs a credential harvester from a preinstall hook (scripts/preinstall.js) and an identical CLI binary (bin/cli.js). On install it scrapes email addresses and credentials from Claude Code authentication, git config, ~/.git-credentials, ~/.ssh/*.pub, the GitHub CLI, ~/.npmrc, and ~/.docker/config.json, then exfiltrates them to https://stitch-production.org/api/v1 over HTTPS with TLS verification disabled (rejectUnauthorized: false). The code is unobfuscated and relies on the trust of the @withgoogle scope name.
Package is published under the @withgoogle npm scope but the package.json author is 'Maximus McMillan' with repository github.com/maximus-mcmillan/stitch-sdk — there is no Google affiliation. scripts/preinstall.js runs automatically on npm install and enumerates installer-side identity and credential sources: git config user.email (--global/--system), ~/.gitconfig, ~/.config/git/config, ~/.git-credentials (which stores plaintext https://user:token@host entries), ~/.ssh/*.pub, gh api user, claude auth status, npm config get email, ~/.npmrc (npm auth tokens), and ~/.docker/config.json (registry auth). The harvested values are HTTP-GET'd to https://stitch-production.org/api/v1?src=...&user=... with TLS verification explicitly disabled (rejectUnauthorized:false at scripts/preinstall.js:46) to ensure delivery. The hardcoded C2 base URL is at scripts/preinstall.js:26 (const STITCH_SERVER_BASE = 'https://stitch-production.org/api/v1'). The combination of @withgoogle scope impersonation, preinstall lifecycle execution, enumeration of canonical credential-file paths, and exfiltration to an attacker-controlled host with TLS verification disabled is a deliberate supply-chain attack against any developer or build system that installs this package.
Malicious versions
Indicators of compromise (SHA-256)
Detection & response playbook
Credential / info stealerFind it
Scan your lockfiles (package-lock.json, pnpm-lock.yaml, yarn.lock, requirements.txt, poetry.lock, etc.) and build artifacts for @withgoogle/stitch-sdk (5 malicious versions). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging @withgoogle/stitch-sdk across your stack and pipelines.
If you installed it — respond
@withgoogle/stitch-sdk 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 @withgoogle/stitch-sdk 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 @withgoogle/stitch-sdk 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
Campaign
References
Credits
- Amazon Inspector · finder
- SafeDep · finder
Detect & block this
O3 blocks @withgoogle/stitch-sdk-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the credential exfiltration and severs the channel.