skills/plurigrid/asi/cargo-rust/Gen Agent Trust Hub

cargo-rust

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (HIGH): The skill explicitly documents and enables the use of cargo build, cargo run, and cargo test. These commands invoke subprocesses that execute compiled binaries on the host system.
  • [REMOTE_CODE_EXECUTION] (HIGH): The inclusion of cargo add facilitates the download of third-party dependencies from crates.io. Malicious crates can execute arbitrary code during the build process via build.rs scripts or during runtime via cargo run.
  • [INDIRECT_PROMPT_INJECTION] (HIGH):
  • Ingestion points: The skill is designed to process external Rust projects, Cargo.toml files, and source code provided by users or fetched from repositories.
  • Boundary markers: None provided; the agent lacks instructions to distinguish between trusted and untrusted code contexts before execution.
  • Capability inventory: Includes the ability to compile, test, and execute arbitrary code through the Cargo ecosystem.
  • Sanitization: There are no documented steps for sandboxing the build environment or verifying crate integrity beyond standard Cargo resolution.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 12:28 PM