Ralphinho Rfc Pipeline

When you've got a feature too large for a single agent pass, this gives you a structured way to break it into independently verifiable work units with explicit dependency management. You write an RFC, decompose it into a DAG of units (each with scope, tests, risk level, rollback plan), then run each through a quality pipeline before hitting the merge queue. The tier system is helpful: isolated file edits are Tier 1, schema or auth changes are Tier 3. The recovery mechanism for stalled units is smart, you evict, snapshot findings, and regenerate with narrower scope. It's opinionated about merge discipline (rebase on integration, rerun tests after each merge) which will save you from mysterious integration failures.

Install to Claude Code

npx -y skills add affaan-m/everything-claude-code --skill ralphinho-rfc-pipeline --agent claude-code