pump-ai-agents
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documents standard agent integration patterns for the Solana ecosystem, providing templates for instruction files and skill registries.
- [COMMAND_EXECUTION]: Defines terminal management rules (isBackground, auto-kill) and MCP server configuration using
npxwith vendor-affiliated packages (@pump-fun/mcp-server). These are standard operational configurations for the intended platform. - [CREDENTIALS_UNSAFE]: The skill explicitly mandates security hardening for sensitive data, including instructions to use
0600file permissions for keypairs and zeroing out key material after use to prevent exposure.
Audit Metadata