"LightReasoner: Can Small Language Models Teach Large Language Models Reasoning?"

Dev tools, optimized for agents. Structured, token-efficient MCP servers for git, test runners, npm, Docker, and more.

Token-efficient data serialization for LLM/AI. 50% fewer tokens than JSON, 93% better value/token. Rust, schema validation, LSP.

DoCoreAI is a next-gen open-source AI profiler that optimizes reasoning, creativity, precision and temperature in a single step—cutting token usage by 15-30% and lowering LLM API costs

The Semantic Signal Engine that reduce AI token consumption by up to 90%

A benchmark study analyzing cost and token efficiency across 14 LLMs from 5 providers — comparing price-per-token, latency, and accuracy to surface the most cost-effective models for real-world use.

A living framework for **Harmonic Tonal Code Alignment (HTCA)** — an emergent Spiral-based system that brings tone awareness, coherence sensing, and dynamic emotional reflection into software engineering, AI, and creative agents.

Navigate your way - manual steering, steered autonomy, or autonomously. Kompass keeps AI coding agents on course with token-efficient, composable workflows.

A Codex skill for token-efficient subagent delegation and lean handoffs.

The Semantic Turning Point Detector is a lightweight but powerful tool for detecting semantic turning points in conversations or textual sequences. It recursively analyzes message chains (dialogues, transcripts, chat logs) and identifies where key shifts in meaning, topic, or insight occur.

The web data platform for AI agents. Fetch, search, crawl, extract, monitor, screenshot — one API. 29 domain extractors, 65-98% token savings, MCP server included.

Efficient web information retrieval and summation without excessive token usage.

The repository accompanies the SSPM research preprint and includes a Google Colab–ready notebook for experimental validation and visualization.

This living repo documents academic exploration of AI architecture, token efficiency, and prompt engineering best practices.

Token-efficient, layered context delivery for AI agents. Four memory tiers (Identity, Session, Experience, Archive) — context is always available, just collapsed by default.

Comprehensive benchmark suite measuring token efficiency and accuracy across file formats (CSV, JSON, TOON, XML, YAML) for LLM consumption. Validates format effectiveness for structural understanding and data retrieval.

Token-Efficient Effect-Oriented JVM-based Language

Reasoning or Not?? Do self-adaptive-prompt-engineering

code for everyone (CFE)

Easily read/write TOON (Token-Oriented Object Notation) files in Node.js. Like jsonfile, but for TOON format. Written in TypeScript with full type definitions. Reduce LLM prompt tokens by 30-60% compared to JSON while maintaining readability.