The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files and technical reference guides. It does not contain any executable scripts, command-line operations, or persistence mechanisms.
[SAFE]: Documentation mentions standard and well-known Rust ecosystem crates, including tokio, sqlx, insta, rstest, proptest, and quickcheck. These are referenced neutrally as part of the technical guidance for testing.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary purpose is to process and review user-provided Rust source code.
Ingestion points: User-provided Rust code files, specifically those matching _test.rs or containing #[cfg(test)] modules.
Boundary markers: None identified in the skill instructions to delimit untrusted code from instructions.
Capability inventory: The skill is limited to generating text-based review reports; it does not contain subprocess calls, file-write operations, or network capabilities.
Sanitization: No evidence of input sanitization or validation is present.
[SAFE]: The reference materials use standard local connection strings (e.g., postgres://localhost:5433/test) as placeholders for testing configurations, which does not involve the exposure of real credentials.

rust-testing-code-review