HomeProjectsArticlesPK BoardAbout
Explore Projects

DeepSeek-Reasonix

Active
GitHub Go MIT

Description

DeepSeek-Reasonix is a DeepSeek-native AI coding agent for the terminal, engineered around prefix-cache stability — leave it running.

Key Features

  • DeepSeek-native AI coding agent designed for terminal use with prefix-cache stability
  • Config-driven with providers, tools, and plugins all declared in reasonix.toml
  • Multi-model composable: DeepSeek flash/pro and MiMo presets with optional dual-model execution
  • Plugin-driven external tools via stdio JSON-RPC (MCP-compatible) subprocesses
  • Zero-friction single static Go binary with cross-compilation to 6 targets
  • Permission system with deny/ask/allow policies and sandbox enforcement for file writes

Use Cases

💡 Terminal-based AI-assisted coding for DeepSeek-optimized workflows
💡 Long coding sessions with low token costs via prefix cache stability
💡 Extensible agent setup with custom plugins and MCP-compatible tool integration
💡 Secure code editing with sandbox enforcement and permission gating
💡 Dual-model workflows combining an executor for fast tasks and a planner for complex reasoning

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (35.4k stars)
  • • Permissive open-source license (MIT)

Tags

coding-agent deepseek cli tui prompt-caching go

Categories

💻 Coding Agent

Quick Start

npm i -g reasonix && reasonix setup && export DEEPSEEK_API_KEY=sk-... && reasonix chat
Visit GitHub Visit Website

Related Projects

CodeWhale

40.9k · Rust
Active A+

CodeWhale is a DeepSeek + MiMo powered terminal coding agent delivering high-performance code generation and reasoning for developers.

coding-agentdeepseekcli +2
  • · Constitution architecture — Nine-tier authority hierarchy where user intent > verification > confidence, ensuring correct decision chains in conflicting contexts
  • · Concurrent sub-agents — Up to 20 parallel sub-agents with async task delegation, results returned inline via completion sentinels
  • · LSP diagnostics integration — Auto-runs rust-analyzer, pyright, typescript-language-server after every edit, injecting diagnostics into next reasoning step
View full details → View on GitHub

oh-my-pi

29.1k · TypeScript
Active A

oh-my-pi is an AI coding agent for the terminal with hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more.

coding-agentclitui +3
  • · Hash-anchored edits for first-attempt-landing precise code modifications, avoiding edit drift
  • · 32 built-in tools + 13 LSP operations + 27 DAP debug operations — batteries included
  • · Deep LSP integration — renames automatically update all references via willRenameFiles
View full details → View on GitHub

Crush

27.9k · Go
Active A

Open-source agentic coding CLI by the Charm team, supporting multiple LLM backends for autonomous coding in the terminal

coding-agentcliterminal +2
  • · Multi-model support — connect to multiple LLMs via OpenAI and Anthropic-compatible APIs including Groq, OpenRouter
  • · Mid-session model switching — switch between different LLM providers while preserving context
  • · LSP-enhanced context — leverage Language Server Protocol for additional code context and project structure understanding
View full details → View on GitHub

Pi

101.0k · TypeScript
Active A+

Pi is a full-stack AI agent toolkit featuring a coding agent CLI, unified LLM API, TUI and Web UI libraries, Slack bot, and vLLM inference pods.

coding-agentclillm +2
  • · Unified multi-provider LLM API — pi-ai package provides unified interface for OpenAI, Anthropic, Google, and other LLM providers
  • · Interactive coding agent CLI — pi-coding-agent provides command-line interactive coding agent with tool calling and state management
  • · Agent runtime — pi-agent-core provides agent runtime with tool calling and state management for multi-step task execution
View full details → View on GitHub