rust-unsafe-ffi

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill establishes best practices for Rust development, focusing on memory safety and correct FFI usage. It references official documentation and well-known community tools from the Rust project.
  • [PROMPT_INJECTION]: The skill involves an attack surface for indirect prompt injection as it processes and generates code based on user inputs.
  • Ingestion points: The agent processes user-supplied Rust source code, FFI declarations, and memory layout descriptions.
  • Boundary markers: The guidelines require clear documentation of safety contracts (# Safety) and localized unsafe blocks, which serve as logical boundaries for code analysis.
  • Capability inventory: The skill relies on standard Rust toolchain commands (cargo test, cargo clippy, cargo miri) for validation.
  • Sanitization: The instructions focus on establishing internal invariants and explicit validation (e.g., null checks, alignment) to prevent memory corruption, which indirectly mitigates the impact of malformed code inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 06:56 PM
Security Audit — agent-trust-hub — rust-unsafe-ffi