The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python audit script (scripts/audit_java_zig_tests.py) and standard Zig build commands (zig build migration-test). These commands are essential for the skill's primary purpose of migration verification and are executed against the user's local project files.
[INDIRECT_PROMPT_INJECTION]: The audit script processes external data including Java source files, Zig source files, and JSON manifest files.
Ingestion points: File reading occurs in extract_java_tests, extract_zig_test_names, and validate_manifest within scripts/audit_java_zig_tests.py.
Boundary markers: None explicitly defined in the prompt instructions.
Capability inventory: The script is limited to reading files, regex matching, and calculating SHA-256 hashes. It does not write to the filesystem (outside of the provided test suite) or perform network requests.
Sanitization: The script implements a safe_file function that explicitly blocks absolute paths and parent directory traversal (..), ensuring the agent only accesses files within the designated project roots.
[DYNAMIC_EXECUTION]: The regression test file scripts/tests/test_audit_java_zig_tests.py uses importlib.util to dynamically load the audit script for unit testing. This is a standard development pattern for local testing and does not introduce runtime risk during normal skill usage by the agent.

zig-java-migration-testing