Show HN: Rocky-Project Hail Mary agent skill that cut output tokens ~47%

Category: devtools

Tags: ai-agents, prompt-engineering, token-optimization

Score: 4.3/10 (Innovation: 4, Technical: 3, Documentation: 6, Utility: 4)

Rocky is an AI agent plugin that makes language models respond in the terse, literal style of Rocky from Project Hail Mary, reducing output tokens by ~47% while preserving technical accuracy. It's interesting as a creative application of prompt engineering that combines literary inspiration with practical token optimization for AI interactions.

Target audience: backend devs

Repository: https://github.com/hpbyte/rocky · Python

View on Hacker News