The Agent Skills Directory

[SAFE]: The skill establishes a structured 'Meta-Cognition Framework' that guides the agent to analyze the entry layer of a query (Mechanics, Design, or Domain) before responding, ensuring technical accuracy and context-awareness.
[SAFE]: The 'Negotiation Protocol' defined in patterns/negotiation.md manages complex or comparative queries by requiring the agent to disclose gaps and confidence levels when synthesizing information from multiple sources.
[SAFE]: The skill promotes security best practices by recommending specific lints (warning on unsafe_code) and integrating with established Rust security tools like cargo-audit, geiger, and miri for code reviews.
[SAFE]: The use of instructional markers such as 'CRITICAL' and 'HIGHEST PRIORITY' in the skill description and instructions are intended for behavioral routing and do not attempt to bypass core safety guardrails or agent constraints.
[SAFE]: No data exfiltration, credential harvesting, or unauthorized network operations were identified. The suggested commands are standard developer tools and utility invocations.

rust-router