Breed better LLM prompts through evolutionary optimization — corpus-evaluated epochs, failure diagnosis, targeted mutations. Agent skill for Cursor, Claude Code, Codex, OpenClaw.
Stars
2
Forks
0
Watchers
2
Open Issues
1
Overall repository health assessment
No package.json found
This might not be a Node.js project
15
commits
scripts: anchor semantic bootstrap to optional narrative core (NARRATIVE_CORE, --core-file)
8af9ce6View on GitHubscripts: add Ollama embedding semantic bootstrap for axis mix
b7e8c58View on GitHubdocs: clarify loop-runner integration (Ralph, Claude, headless)
0dffc1bView on GitHubnarrative: complete 10-cycle NDD evolution via Ralph Loop (5.5 → 9.0)
25d4facView on GitHubrebrand: Define a problem. Aggregate all best practices. Hack it.
1d7de84View on GitHubfeat: integrate subagent-harness as candidate format for Run/Re-run
1e096a6View on GitHubdocs: redesign README based on readme-architect skill principles
af21f17View on GitHub