skills/sablier-labs/solsab/solana/Gen Agent Trust Hub

solana

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a just task runner to automate development workflows. Commands like just build, just codegen, and just test are standard for compiling Rust programs, generating TypeScript bindings from IDLs, and running the test suite.
  • [EXTERNAL_DOWNLOADS]: The documentation provides links to official external resources for the Solana ecosystem, including Anchor, Metaplex Core, and Trident documentation. These links point to official domains (e.g., anchor-lang.com, solana.com, metaplex.com) to ensure developers have access to the latest verified documentation.
  • [SECURITY_BEST_PRACTICES]: A significant portion of the skill is dedicated to security education. It explicitly details patterns for avoiding common Solana vulnerabilities such as missing signer/owner checks, integer overflows, and PDA seed collisions. It mandates the use of Anchor's built-in security features and checked arithmetic.
  • [TESTING_GUIDANCE]: The skill integrates robust testing frameworks, including Vitest for unit/integration tests and Trident for property-based fuzzing. It provides detailed templates for time-travel testing and invariant validation, which are critical for financial applications on Solana.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 08:14 PM
Security Audit — agent-trust-hub — solana