https://github.com/esengine/DeepSeek-Reasonix
DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.
https://github.com/esengine/DeepSeek-Reasonix
agent agent-framework ai-agent ai-coding cli coding-agent deepseek developer-tools ink llm prompt-caching r1 terminal tool-use tui typescript
Last synced: about 2 months ago
JSON representation
DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.
- Host: GitHub
- URL: https://github.com/esengine/DeepSeek-Reasonix
- Owner: esengine
- License: mit
- Created: 2026-04-21T08:27:02.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2026-05-30T03:05:20.000Z (about 2 months ago)
- Last Synced: 2026-05-30T03:11:05.061Z (about 2 months ago)
- Topics: agent, agent-framework, ai-agent, ai-coding, cli, coding-agent, deepseek, developer-tools, ink, llm, prompt-caching, r1, terminal, tool-use, tui, typescript
- Language: TypeScript
- Homepage: https://esengine.github.io/DeepSeek-Reasonix/
- Size: 15.9 MB
- Stars: 13,931
- Watchers: 51
- Forks: 807
- Open Issues: 362
-
Metadata Files:
- Readme: README.ja-JP.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- Funding: .github/FUNDING.yml
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
- awesome-repo - DeepSeek-Reasonix - native AI coding agent for your terminal (Tool / Vibe coding tool)
- awesome-cli-agents - esengine/DeepSeek-Reasonix - DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running. [Terminal] (19k⭐) (🚀 AI Tools for Vim, Neovim, and Terminal)
- awesome-ai-agents - esengine/DeepSeek-Reasonix - DeepSeek-Reasonix is a DeepSeek-native AI coding agent for the terminal, designed for low-cost, long-session use with prefix-cache stability and a config-driven, multi-model architecture. (AI Agent Frameworks & SDKs / Multi-Agent Collaboration Systems)
- StarryDivineSky - esengine/DeepSeek-Reasonix - Reasonix` 是一款专为开发者打造的终端原生 AI 编程助手,旨在彻底解决传统命令行 AI 工具在长周期代码迭代中因上下文丢失、响应延迟高及会话不稳定所导致的开发中断痛点。该项目的核心优势首先体现在其深度契合 DeepSeek 模型特性的原生架构设计,摒弃了冗余的中间抽象层,使大模型的逻辑推理能力得以在终端环境中实现零损耗调用;其次,项目独创的前缀缓存稳定机制如同为对话记忆安装了“防抖阻尼”,能够智能复用历史交互的已生成内容,从而在长时间运行中显著降低 API 消耗并保障响应速度的连贯性;最后,其纯终端优先的设计理念无缝契合开发者现有的工作流,无需依赖臃肿的 Web UI 或额外部署服务即可通过 SSH、Tmux 等环境实现沉浸式辅助编程。从技术原理来看,该项目的工作逻辑可类比为一位始终记得你所有草稿的资深结对程序员,当你在终端输入指令时,系统并非每次都从零开始构思,而是将当前代码库与历史对话共同编译为一份“动态前缀索引”,随后调用底层推理引擎仅针对新增需求生成增量代码块;这种机制通过流式传输实时渲染至终端界面,并利用缓存池锁定已确认的上下文片段,使得复杂的多轮调试过程如同在一条不断向前延伸的轨道上平稳推进,既避免了重复计算带来的资源浪费,又确保了项目状态在长时间运行中的无缝续接。 (A01_文本生成_文本对话 / 大语言对话模型及数据)
- awesome-opensource-ai - DeepSeek-Reasonix - DeepSeek-native AI coding agent for the terminal, designed around prefix-cache stability. MIT licensed.  (4. Agentic AI & Multi-Agent Systems)
- awesome-local-llms - DeepSeek-Reasonix - native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running. | 23,452 | 1,416 | 626 | 0 | 59 | MIT License | 0 days, 1 hrs, 56 mins | (Open-Source LLM & Agent Projects)
- awesome-harness-engineering - DeepSeek-Reasonix - native coding agent harness engineered around prefix-cache stability as a loop invariant: immutable-prefix / append-only-log / volatile-scratch partitioning achieves 99.82% cache-hit rates and ~5× cost reduction on long sessions. The most detailed public case study of designing an entire agent loop to preserve a provider-specific economic property rather than treating caching as an afterthought.  (Reference Implementations / Demo Harnesses)
- awesome-ai-200 - esengine/DeepSeek-Reasonix
- awesome-ai-tools - Reasonix - native coding agent, for your terminal. | (🖥️ AI-Native IDEs & Editors)
- awesome-agent-harness - Reasonix - 05 | esengine | Yes | (Agent Harness 🚀 / 5. The Convergence: Harness-as-a-Service (HaaS))
- awesome-ai-coding - DeepSeek-Reasonix - Reasonix?style=flat-square&label=%E2%98%85)](https://github.com/esengine/DeepSeek-Reasonix) | esengine | DeepSeek-native terminal coding agent. | Engineered around prefix-cache stability, low input-token overhead. | (🤖 AI Coding CLI Agents & SWE Systems / 2026-07-12 — Weekly Update)