Cyber Security News

Weaponized PyPI Package Executes Supply Chain Attack to Steal Solana Private Keys

A chilling discovery by Socket’s Threat Research Team has exposed a meticulously crafted supply chain attack on the Python Package Index (PyPI), orchestrated by a threat actor using the alias “cappership.”

The attack leverages a malicious package named semantic-types to deploy a covert key-stealing payload, specifically targeting Solana blockchain developers.

This campaign, which has already seen over 25,900 downloads across six implicated packages, embeds a backdoor that silently exfiltrates private keys, posing a severe risk to developer environments and CI/CD pipelines.

At the time of reporting, these packages remain live on PyPI despite efforts to have them removed.

Sophisticated Malware Targets Blockchain Developers

The core of this attack lies in semantic-types, a package that serves as the delivery mechanism for the malicious payload.

Five other packages solana-keypair, solana-publickey, solana-mev-agent-py, solana-trading-bot, and soltrade depend on semantic-types, ensuring that a single pip install command for any of these libraries automatically pulls and executes the hidden malware.

solana-mev-agent-py malicious package

Once installed, the payload employs monkey patching, a dynamic Python technique, to alter Solana key-generation methods in the solders library at runtime.

This allows the malware to intercept private keys during creation without modifying the original source code.

The stolen keys are then encrypted with a hardcoded RSA-2048 public key, encoded in Base64, and embedded in spl.memo transactions on Solana Devnet for retrieval by the threat actor.

This innovative use of blockchain as an exfiltration channel bypasses traditional network defenses, as the traffic appears as routine wallet activity, making detection extraordinarily challenging.

Stealthy Exfiltration via Transitive Dependencies

According to Socket Report, the threat actor’s sophistication extends beyond technical execution to social engineering.

Polished README files, strategic links to legitimate Stack Overflow posts, and imitation of trusted Solana tooling names create an illusion of credibility, encouraging unsuspecting developers to adopt these packages.

The campaign timeline reveals a deliberate strategy: initial benign releases of semantic-types and related packages in December 2024 built trust, only for malicious updates starting January 26, 2025, to introduce the payload, expanding the attack’s reach with additional releases through early February.

semantic-types

This delayed payload delivery, combined with transitive dependencies, underscores a calculated effort to maximize impact while minimizing suspicion.

Developers who installed these packages post-January 26, 2025, must assume all Solana private keys on affected systems are compromised.

The on-chain nature of the exfiltration means historic keys can be replayed by the attacker to drain funds whenever wallets are active on mainnet.

Security recommendations include deep dependency inspection, stricter CI/CD controls, and continuous scanning for suspicious runtime behaviors like unauthorized cryptographic operations.

Tools like Socket’s GitHub app and browser extension can help detect hidden execution paths during package installs.

Indicators of Compromise (IOCs)

CategoryDetails
Malicious PyPI Packagessemantic-types, solana-keypair, solana-publickey, solana-mev-agent-py, solana-trading-bot, soltrade
Threat Actor Identifierscappership (PyPI alias), cappership@proton[.]me (email), D782zqWjgSvy4hQoqzY1ySrGrotnXm1suJeXFur8sAko (Solana public key)
RSA Public Key Fingerprint5a4d8480c9d1e82ba102f200258882fb9e694e8fc0343b6982c5540beccdca62 (SHA-256)

Find this News Interesting! Follow us on Google News, LinkedIn, & X to Get Instant Updates!

Aman Mishra

Aman Mishra is a Security and privacy Reporter covering various data breach, cyber crime, malware, & vulnerability.

Recent Posts

New GhostMap Android Vulnerability Can Crash Phones Permanently Without Opening an App or Granting Permissions

Bitdefender researchers have revealed a vulnerability called GhostMap that affects Android devices. This vulnerability can…

6 minutes ago

Microsoft Paint Vulnerability Allows Hackers to Execute Malicious Code Remotely Without User Interaction

Microsoft has announced a vulnerability in Windows Paint that could let unauthenticated attackers execute malicious…

29 minutes ago

ARTEX AI Framework Linked to Bank Breaches and Cloud Attacks Found Exposed Across Hundreds of Servers

An autonomous penetration-testing framework linked to South Korean banking breaches and cloud supply-chain attacks is…

59 minutes ago

Shopify Makes IP Allowlisting Mandatory for All Public Apps to Prevent Stolen API Token Abuse

Shopify is making IP allowlisting mandatory for all public apps to limit abuse of stolen…

1 hour ago

OpenAI’s New GPT-6 Cyber Finds 81 High-Severity Firefox Vulnerabilities and 948 Linux Kernel Exploits

OpenAI has introduced GPT-6 Cyber, a security-focused model built to find and validate exploitable vulnerabilities…

1 hour ago

OpenMatter Network Positions Decentralized Computing as an Alternative to Data Centers for AI

Melbourne, Florida, November 5th, 2026, CyberNewswire OpenMatter Network, a pioneer in decentralized infrastructure for agentic…

13 hours ago