Meta Cognition Parallel

This is an experimental approach to debugging Rust problems by running three parallel analyses: language mechanics (ownership, borrowing), design patterns (Arc vs Rc, when to clone), and domain constraints (FinTech safety requirements, web scalability needs). It detects whether you have the agent files installed and either spawns three concurrent sub-agents or runs the layers sequentially inline. The synthesis step is where it gets interesting, connecting how domain requirements force specific design choices which then dictate language features. Best for complex architectural questions where the "right" answer depends on context you haven't fully articulated yet. The parallel execution is genuinely experimental and might time out on heavy queries.

Install to Claude Code

npx -y skills add actionbook/rust-skills --skill meta-cognition-parallel --agent claude-code