solana-agent-kit

Warn

Audited by Socket on Sep 15, 2026

2 alerts found:

Securityx2
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill is purpose-aligned and uses mostly legitimate same-project npm distribution, but it enables autonomous blockchain transactions and forwards a raw wallet private key into agent/MCP processes. The main risk is real-world financial autonomy plus sensitive-key handling, not hidden malware or deceptive exfiltration.

Confidence: 91%Severity: 76%
SecurityMEDIUM
examples/autonomous-agent/README.md

No clear indicators of classic malicious code (obfuscation, backdoors, eval/Function-based payloads, suspicious exfiltration, or system tampering) are present in the provided fragment. However, the module is inherently high security risk because it loads a real Solana private key, enables LLM-driven autonomous/event-triggered on-chain actions via Solana tools, and the safety guardrails shown appear not to be enforced in the primary execution flows. The likely risk is unintended or attacker-influenced financial actions (e.g., prompt/trigger-driven trades), rather than covert malware.

Confidence: 62%Severity: 78%
Audit Metadata
Analyzed At
Sep 15, 2026, 05:46 AM
Package URL
pkg:socket/skills-sh/sendaifun%2Fskills%2Fsolana-agent-kit%2F@c6f7ebdcf171784790c5f1d0dfe48d1512c44f8092d0c26affede19662bcb175