Jason5330/ai-memory-system
Self-evolving memory for Claude Code & Codex — two-layer (personal + per-project), reflects into doctrine, harvests repeated workflows into skills, hard-blocks broken tools.
Self-evolving memory for Claude Code & Codex — two-layer (personal + per-project), reflects into doctrine, harvests repeated workflows into skills, hard-blocks broken tools.
npx add-skill Jason5330/ai-memory-systemSelf-evolving memory for Claude Code & Codex — two-layer (personal + per-project), reflects into doctrine, harvests repeated workflows into skills, hard-blocks broken tools.
Five Claude Code skills that drive a 72-96h mathematical-modeling contest (MCM/ICM, CUMCM, 等) end-to-end: triage → modeling → solving → writing → submit.zip. State-on-disk, resumable across crashes, integrity-gated.
Custom skills for agents to support academic paper writing and thesis management
Stop writing 200-line skills that Claude skims. Convert any skill into a gated micro-pipeline with hard YES/NO quality checks.
occams-agentic: harness-agnostic AI agent framework. Universal skills, scripts, conventions, and knowledge management patterns that work with any AI coding harness
Best static AI text humanizer. Two research-grounded skills that work in any LLM (Claude, ChatGPT, Gemini, Codex): humanize beats perplexity-based detectors, ai-check produces forensic scoring with evidence-quoted flags. Nine levers, 50+ peer-reviewed sources, 2024-2026 detection literature.
Demo RAG API (FastAPI, OpenAI, ChromaDB, Docker) automatically generated using the OpenAI Codex CLI tool. Highlights Codex's capability for rapid, complex application development.