The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the review and testing of Rust macro code, creating a surface for indirect attacks through malicious input code.
Ingestion points: The agent processes Rust source code, macro definitions, and test suites provided by the user or retrieved from a repository (SKILL.md).
Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent to disregard instructions or malicious logic embedded in the code being analyzed.
Capability inventory: The skill utilizes shell commands for verification, specifically cargo expand, cargo test, and cargo clippy --all-targets (SKILL.md).
Sanitization: Absent. The skill does not specify validation or isolation of the code before executing it via cargo test, which can execute arbitrary code defined in the crate's build.rs or test modules.

rust-macros