Show HN: Forge - Guardrails take 8B model from 53% to 99% on agentic tasks

Forge transforms a small 8B model from 53% to 99% accuracy on agentic tasks using structured output enforcement and step-level verification — one of the week's most technically engaged threads at 174 points. The technique achieves near-perfect execution without weight changes, confirmed across 47 diverse tasks, offering a lightweight, cost-effective alternative for real-world deployment.