rust-web-backend

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill references the use of the DATABASE_URL environment variable for managing database connection strings. This is a standard and safe practice for secret management, provided the environment is configured correctly.- [INDIRECT_PROMPT_INJECTION]: The skill guides the implementation of web handlers that ingest untrusted data via axum extractors (Json, Path, Query). It mitigates risks through:
  • Ingestion points: Data enters via typed extractors in architecture.md and templates/rest-service.md.
  • Boundary markers: The use of typed Rust structs and explicit Result<T, AppError> return types provides strong boundaries for handling external input.
  • Capability inventory: The code interacts with databases and network sockets as part of its primary function.
  • Sanitization: The skill mandates the use of sqlx::query! macros which utilize bind parameters to prevent SQL injection attacks.- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill references well-known and trusted sources within the Rust ecosystem, such as official documentation at tokio.rs and the standard package registry crates.io, for its dependency stack.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 04:06 PM
Security Audit — agent-trust-hub — rust-web-backend