Search

DeepSeek Harness 官方架构的 Windows 桌面发行版 (Desktop distribution of the official DeepSeek Harness)

A Framework for Agents work efficiency

DeepSeek Harness Plus: timely fixes for upstream bugs, early features, practical extensions, and curated presets.

A lightweight desktop client for DeepSeek Harness.

A lightweight Python framework for building OpenAI Agents with pluggable model providers.

A self-evolving multi-agent AI system that autonomously writes, sandboxes, evaluates, and improves its own agents with an LLM router, event-driven brain, and a 3D web console.

A local-first agentic coworker you can configure without limit. Choose your files and model. Walk away with a spreadsheet, deck, or working app.

AI-driven Abaqus workbench: describe a model in a spec, get a real solved one back. Every capability is backed by a check you can run on your own seat, and the gates refuse a doubtful answer instead of returning it. AGPL-3.0 + commercial licence.

Watch your Claude Code subagents work in a pixel-art office

Server-side helpers for @infra-tools/agentic-ui — generic Agent interface, AG-UI SSE route handler, in-memory thread store.

Optional runtime plugin that gates ToolRegistry / ComponentRegistry reads via OPA decisions (Open Policy Agent). Slice OPA-B / ADR-040. Stays outside the core lib per ADR-010 D4.

An AI agent that reads research PDFs (via mupdf.js) and extracts field, method, result, and impact — with compare, benchmark, methodology-figure, and code-generation commands.

OpenCode Self-Improving Orchestration Plugin - An adaptive control layer for continuous self-evaluation, workflow improvement, and persistent project intelligence

Ultra-lightweight, scalable CLI/TUI coding agent optimized for tiny local models & remote runtimes

Pragma — an agentic AI coding harness. Downloads and launches the latest release for your platform.

Overleaf, but the collaborator is your coding agent: point at the PDF, it edits the LaTeX and rebuilds — all local.

Ness Agent is an experimental, hackable agent harness.

Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...

Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio, with 16B active parameters out of 280B total. It is the lightest model in the Dots 3 family and is...

No description available