-
Loop Engineering — The Shift from Prompting to Designing Autonomous Agent Systems
The engineering community is moving beyond turn-by-turn prompting. Loop engineering is the practice of designing self-sustaining systems that discover, delegate, verify, and iterate — without you in the driver's seat.
-
llm-context-fmt — Your Prompt's Format Choice Changes the Answer
New CLI tool reveals how format constraints (JSON, XML, Markdown, plain text) systematically bias LLM outputs — based on real July 2026 ArXiv research.
-
readme2skill — Auto-Generate AI Agent Skills from Any Project README
CLI that reads your README, docs, and config to instantly produce SKILL.md — making any project usable by Claude Code, Codex, Cursor, and Gemini CLI without manual skill authoring.
-
vuln-seeker — CLI Code Security Scanner with AI Triage
A lightweight CLI tool that scans your codebase for common security vulnerabilities using regex patterns, with optional AI-powered analysis for deeper triage.
-
model-cards — Browse and Compare Open-Weight AI Models from Your Terminal