Coding agentSoftware developmentLanguage Server Protocol
crush
@charmbracelet·Go
Works on code in the terminal while switching models, preserving sessions, and connecting MCP tools.
Project overview
Crush is a terminal coding assistant that can switch models during a session while preserving context and keeping multiple project sessions. It uses language servers for additional code context and connects MCP servers over stdio, HTTP, or SSE. It supports many OpenAI- and Anthropic-compatible providers as well as local models; users need to configure a working provider, credentials, or local endpoint before sending model requests.
Use Gemini from the terminal to inspect code, make changes, and automate development tasks.
Project overview
Gemini CLI lets developers work with Gemini directly from a terminal to inspect repositories, generate or edit code, debug issues, and run tool-assisted workflows. It includes file, shell, web, and search tools, supports MCP extensions and non-interactive scripting, and offers Google sign-in, API key, or Vertex AI authentication. A personal Google account can use the documented free tier.
Build controllable AI agents in Python or TypeScript
Project overview
Strands Agents is a monorepo with Python and TypeScript SDKs for building model-driven agents. The SDKs support tools, MCP, streaming, multi-agent patterns, hooks, guardrails, context management, and multiple model providers, with examples for local development and deployment. Python requires 3.10 or newer and the TypeScript package requires Node.js 20 or newer; the default quickstart uses Amazon Bedrock credentials.
Run a Rust coding agent with sandboxing and many model providers
Project overview
VT Code is a Rust coding agent for long-running automation, available through an interactive TUI or headless commands and compatible with more than 21 cloud and local model providers. Restricted shell sandboxing, tool guardrails, and audit logs define its execution boundaries, while worktree isolation and Skills/MCP support extend workflows. Local inference through Ollama, LM Studio, and llama.cpp is still marked experimental.
Calls Claude Agent from Python with async queries, in-process tools, hooks, and interactive sessions.
Project overview
Claude Agent SDK for Python exposes an async query function and ClaudeSDKClient for bidirectional, interactive conversations with Claude Code. Custom tools and hooks can run as in-process MCP servers, while external MCP servers can be used alongside them. The package bundles the Claude Code CLI by default, requires Python 3.10 or newer, and allows a custom CLI path.
Route coding-agent work through replayable, evidence-gated multi-agent lanes
Project overview
OMK is a provider-neutral control plane for Codex, Claude Code, OpenCode, and local coding agents. It routes work into scoped DAG lanes, verifies and measures execution, exposes MCP and context-budget state, and records artifacts that can be replayed; the monorepo also ships an interactive CLI, agent runtime, multi-provider LLM API, and operator TUI. OMK extensions and MCP servers support workflow integrations such as Slack or chat automation.
Runs Kimi’s terminal agent with shell mode, MCP servers, and IDE integrations.
Project overview
Kimi CLI combines a terminal agent with a Ctrl-X shell mode, a VS Code extension, ACP support for compatible editors, Zsh integration, and MCP server management. The project is being replaced by the next-generation Kimi Code CLI, which migrates configuration and sessions during installation. Existing installations remain available, but new users seeking ongoing updates should choose Kimi Code CLI.
Run long coding tasks through an editable local agent harness with readable traces.
Project overview
Agent AFK provides an Apache-licensed harness whose prompts, gates, routing, skills, traces, and provider settings are editable. It supports terminal chat, headless background runs, Telegram notifications, orchestration commands, and migration of plugins or MCP servers from Claude Code or Codex. Node 22 is required, and optional remote notifications need their own credentials.
Run a persistent personal agent with sandboxed tools, memory, and multiple channels.
Project overview
Moltis is a self-hosted personal-agent server written in Rust that connects to multiple model providers and stays available through the web, Telegram, Signal, Discord, Teams, and voice. Tool execution can be isolated with Docker, Apple Container, or WASM, while memory, MCP, scheduling, authentication, and an encrypted vault are built in. Release installs run as a single Rust binary on hardware you control.
CLI appAI and agent developmentModel Context Protocol
mcp-gateway
@docker·Go
Runs isolated MCP servers behind a Docker CLI gateway for multiple AI clients.
Project overview
Docker MCP Gateway is a Docker CLI plugin that can run with Docker Desktop or independently. It isolates MCP servers in containers, organizes them with profiles and OCI catalogs, and gives clients such as VS Code, Cursor, and Claude Desktop one gateway for tools, prompts, and resources. Secrets, OAuth, discovery, and call tracing are supported; Docker Desktop users need version 4.59 or newer with the MCP Toolkit.
Learn AI coding-agent architecture through a manageable CLI project
Project overview
Mini-Kode is an educational coding assistant that presents modern agent architecture in a compact but complete terminal project. Its code shows how tools, model calls, permission approval, streaming, and the agent loop fit together, with file operations, search, command execution, and MCP integration available for developers to study and modify.
Handles interactive or automated terminal development with multiple models, subagents, MCP, and ACP.
Project overview
Kode is an open-source terminal assistant with multi-model switching, expert consultation, specialized subagents, code editing, command execution, MCP extensions, AGENTS.md discovery, and an ACP server mode. It supports interactive, non-interactive, and plan workflows, plus plugins and model configuration files. Its default mode skips most approval prompts, so safe mode is better for sensitive work. The repository also says a major maintenance and refactoring phase runs from July through December 2026.
MCP client or hostAI and agent developmentModel Context Protocol
tuui
@AI-QL·TypeScript
Connect multiple LLM providers and run MCP tools from a local desktop workspace.
Project overview
TUUI is a desktop LLM chat and experimentation workspace built around the Model Context Protocol. It lets users configure hosted or self-managed model endpoints and work with MCP tools, prompts, resources, sampling, elicitation, registry discovery, and MCP bundles. The application is available as a direct download, but MCP use requires a model backend with tool calling and the appropriate Node.js, Python/uv, or Docker runtime for each configured server.
Agent runtimeAI and agent developmentAgent Client Protocol
octomind
@Muvon·Rust
Runs specialist agents with intent-driven context, adaptive sessions, budgets, and policy scripts.
Project overview
Octomind is a Rust agent runtime that runs packaged specialist roles and can activate skills, MCP servers, or sub-agents during a session. It uses adaptive compaction to preserve important context, tracks costs with request and session caps, and enforces deterministic guards, hooks, and validators from scripts. The same binary supports interactive CLI, JSONL, daemon, WebSocket, and ACP modes.
LibraryAI and agent developmentModel Context Protocol
ToolRAG
@antl3x·TypeScript
Retrieve only the most relevant tools for an LLM query
Project overview
ToolRAG registers tools from MCP servers, embeds their descriptions, and uses semantic retrieval to select the functions relevant to a user query. The selected definitions can be passed in an OpenAI-compatible format and executed against the corresponding servers, while LibSQL stores the tool index and embeddings. It is aimed at assistants that need access to many APIs without sending every tool definition on every request.
Analyze, debug, and review code with multiple model providers
Project overview
Binharic is a terminal coding agent with a Tech-Priest persona and support for models from OpenAI, Google, Anthropic, and Ollama. It uses built-in tools, keyword-based retrieval, predefined development workflows, and MCP extensions to handle common coding tasks. The project is still in early development, so bugs and breaking changes are expected.
FrameworkAI and agent developmentModel Context Protocol
agentsilex
@howl-anderson·Python
Builds a small, readable agent framework with swappable models, tools, sessions, and handoffs.
Project overview
AgentSilex is a roughly 300-line Python framework built around readable Agent, Runner, Session, and tool abstractions. LiteLLM lets it switch among more than 100 model providers, while the framework adds tool calls, session history, multi-agent handoffs, MCP clients, streaming, and structured output. OpenTelemetry tracing is included for inspecting execution, and the package requires Python 3.12 or newer.