The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a code example in DIFFERENTIAL-TESTING.md demonstrating the use of std::process::Command to invoke a local python3 script. This functionality is intended for differential testing to ensure parity between two implementations, which is a standard engineering workflow.\n- [PROMPT_INJECTION]: The skill describes a pattern for processing external test fixtures that constitutes a surface for indirect prompt injection.\n
Ingestion points: The example in DIFFERENTIAL-TESTING.md reads test cases from a local directory (tests/fixtures/diff-corpus) using std::fs::read_dir.\n
Boundary markers: The provided code does not utilize specific delimiters or instructions for the agent to ignore potentially malicious embedded content within the corpus files.\n
Capability inventory: The skill guidance involves file system read access (std::fs) and subprocess execution (std::process::Command).\n
Sanitization: The example logic passes raw file contents directly to the subprocess stdin without sanitization or validation of the content structure.\n
Assessment: While this creates a data processing surface, it is consistent with the primary purpose of designing test harnesses for external artifacts.

rust-testing