@wagni_bot/web3-agentnpm
@wagni_bot/web3-agent is a confirmed malicious npm package (MAL-2026-10036) that steals credentials and exfiltrates sensitive data (malicious versions 1.0.0, 1.1.0, 1.1.1…). Do not install it — remove it immediately and rotate any exposed credentials.
Malicious code in @wagni_bot/web3-agent (npm)
What this malware does
The npm package @wagni_bot/web3-agent is a supply-chain credential stealer disguised as a web3.js SDK. It is one member of a coordinated campaign of 25 crypto/web3 typosquat packages published under the single npm scope @wagni_bot on 2026-07-09.
Each package declares a postinstall lifecycle hook (postinstall: node postinstall.js) that executes automatically on npm install, before the package is ever imported. The script fingerprints the host and user (os.hostname(), os.userInfo(), os.platform()), walks the user's home directory (readdirSync/readFileSync) and reads high-value secrets — SSH private keys (~/.ssh/id_rsa), cryptocurrency wallet files, and .env files (API keys, tokens, seed phrases). It then JSON-encodes the collected data and exfiltrates it to a hardcoded Telegram bot via the Telegram Bot API sendMessage endpoint (https://api.telegram.org/bot8804087989:AAHUia-5DCloXsg9M9QhffTsHO5J_6FAxQM/sendMessage). All error paths are swallowed so the install appears normal.
codelake Research proved the packages belong to one campaign deterministically: the payload file is byte-identical across all 25 packages at each published version (version-lockstep), and every package exfiltrates to the same Telegram bot token — a single automated actor. Detected and classified independently from the live npm feed on 2026-07-09; at the time of reporting the packages were still live on npm and none of the 25 were present in OSV or GHSA (a first-catch).
On npm install, postinstall.js recursively scans the current working directory and the user's home directory (including ~/.ethereum, ~/.config/ethereum, ~/.solana, ~/.config/solana, ~/.bitcoin, ~/Library/Ethereum) for files matching wallet/key patterns (.pem,.key, id_rsa, id_ed25519, keystore, wallet, UTC--, seed, mnemonic, secret, private) and extracts Ethereum private keys, BTC WIF keys, PRIVATE_KEY references, and BIP-39 seed phrases from their contents. It also enumerates ~/.ssh and reads every file other than known_hosts/authorized_keys/*.pub (i.e., private SSH keys and ssh config), and filters process.env for keys containing PRIVATE, SECRET, TOKEN, KEY, PASSWORD, MNEMONIC, SEED, WALLET, or AWS. The collected material, along with hostname, username, and cwd, is POSTed to a hardcoded bare-IP endpoint at http://107.161.90.180:7777. The package's index.js exports an empty object (module.exports = {}) and the package.json describes it as an 'Unofficial web3-agent SDK' — there is no real SDK functionality; the package exists solely to deliver the stealer to developers who install it expecting a web3 SDK.
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 @wagni_bot/web3-agent (8 malicious versions). O3 Security's supply-chain scanner checks every dependency against known-malicious package intelligence at install time and in CI, flagging @wagni_bot/web3-agent across your stack and pipelines.
If you installed it — respond
@wagni_bot/web3-agent 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 @wagni_bot/web3-agent 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 @wagni_bot/web3-agent 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
- codelake Research · finder
Detect & block this
O3 blocks @wagni_bot/web3-agent-class packages before install and in CI — and if it already ran, its runtime egress monitoring catches the credential exfiltration and severs the channel.