The Agent Skills Directory

[SAFE]: The skill serves as a safety-oriented workflow definition for Rust programming. It establishes deterministic routing rules for undefined behavior (UB) and ownership management.
[COMMAND_EXECUTION]: The skill mentions various Cargo and Rust toolchain commands (e.g., cargo clippy, cargo test, miri, loom). However, these are listed as 'observed-only evidence' or 'gate commands' that the external executor (not the AI agent itself) must run. The skill explicitly states 'Do not run cargo, Miri, sanitizers, or any toolchain from OMH core.'
[REMOTE_CODE_EXECUTION]: There are no patterns of remote code execution. All toolchain references are local to the user's development environment and managed by the human executor or CI system.
[DATA_EXFILTRATION]: No network operations or attempts to access sensitive file paths (like .ssh or .aws) were found. The skill focuses entirely on code structure and safety gates.
[PROMPT_INJECTION]: The skill contains instructional markers like 'IMPORTANT' and 'CRITICAL' within its safety rules, but these are used legitimately to enforce the UB escalation workflow and safety discipline, rather than to bypass system filters.
[OBFUSCATION]: No obfuscated URLs, base64-encoded commands, zero-width characters, or homoglyph attacks were detected. The content is clear and uses standard technical documentation practices.

omh-rust