Explore the directory

AxisAgentic is an extensible runtime for long-horizon AI agents that also records the trajectories those agents produce, so the same execution traces can support replay, benchmark evaluation and supervised fine-tuning data export.

• Agent frameworks
• AI evals

tty7 is a Rust terminal emulator whose sessions live in a background server rather than the window, so shells, SSH connections and supported coding-agent sessions keep running after the window closes or the machine restarts.

• Terminals

MassGen is a terminal-based multi-agent system that runs several AI models on the same task in parallel, has them critique and build on each other's answers over repeated rounds, and lets the agents vote on the best result.

• Agent frameworks

LatentMAS lets multiple AI agents collaborate by passing latent representations through a shared working memory instead of exchanging text, aiming to reduce token usage and speed up multi-step reasoning.

• Agent frameworks
• Agent memory

A self-hosted, real-time task manager for AI agents that lets a human control and review their own Claude, Codex or other ACP-compatible agents from mobile, web or desktop through the Model Context Protocol.

• Coding agents

A companion code repository for a paid Chinese-language book and course on engineering practices with Claude Code, with a runnable project for each lecture covering memory, subagents, hooks, MCP and SDK usage.

• Tutorials

A 2-bit quantized language model, and the training pipeline for it, small enough to run as a 40KB CP/M binary on a Z80 processor with 64KB of RAM.

• Models

A high-performance operator library for large language models, implemented in the Triton language, that registers with PyTorch's ATen backend to accelerate training and inference across many hardware platforms.

• Models
• Inference

An integration layer that connects AI coding agents to Tencent CloudBase as a backend, installing the stack through a plugin, steering code generation with skills, and operating databases, functions and deploys through MCP.

• MCP servers
• Backend

A Python framework for end-to-end fine-tuning, inference and distillation of RAG retrieval models, covering embedding, ColBERT-style late-interaction, and reranker models.

• RAG

A curated, evidence-backed collection of use cases, prompts, integrations, evaluations and safety notes for OpenAI's GPT-6 Astra model, grounded in OpenAI's own documentation.

• Curated lists

A multi-user, self-hosted personal AI agent platform built on Spring Boot, with multi-agent orchestration, MCP, skills and memory, automatic provider failover, and multi-channel bots such as Telegram, Feishu and DingTalk.

• Agent frameworks

An open protocol that standardizes how AI agents perceive real-world data, so that installing a sensor gives an agent structured, real-time signals in a common schema.

• Agent frameworks

A C++ kernel for building AI agents that run entirely on-device, with sub-100ms latency on Qualcomm NPUs, crash recovery, formal verification, and no cloud dependency.

• Agent frameworks

A local CLI toolchain that indexes months of existing coding-agent session history on a developer's machine so it can be searched, blamed to code, and cut down before feeding it back to an agent.

• Coding agents
• Agent memory

A real-time control room for observing and steering simulated societies of language-model agents, letting a user pause time, question a resident, change what happens next, and replay or export a run.

• Agent frameworks
• AI evals

A course and full source code for building a smaller version of vLLM from scratch in C++ and CUDA, covering model loading, CUDA kernels, KV caching, batching and PagedAttention.

• Inference
• Tutorials

A shared workspace that lets a team open, watch and dispatch work to the coding agents and machines its members already use, from desktop, mobile, web or the CLI.

• Coding agents

A reproducible security evaluation workflow for generative AI, with one product testing an LLM or API's behavior under adversarial scenarios and a second extending that evaluation to a running agent's tool calls and file changes.

• AI evals
• Security

A lightweight WhatsApp bot framework with multi-session support, a plugin system and group management tools, rewritten from the original Raganork project.

• Workflow automation

Open research code exploring the Engram memory-injection architecture as an alternative to LoRA for fine-tuning language and vision models, extended from Qwen to Stable Diffusion.

• Models

A CLI and MCP tool that turns PDFs, Office files, scans, screenshots, charts and diagrams into AI-ready Markdown, using any OpenAI-compatible multimodal model instead of a fixed OCR pipeline.

• MCP servers
• RAG

A spec-driven coding harness that adds persistent context memory, 15 agents and 33 skills to Claude Code and Codex, running a 7-phase plan-first workflow with self-healing check loops.

• Coding agents
• Agent skills

The official Model Context Protocol server for MongoDB, letting an AI agent connect to a self-managed MongoDB deployment or MongoDB Atlas through a common set of tools.

• MCP servers
• Databases