continue

⚡️ 10x - Up to 20x faster AI coding with multi-step Superpowers. Open-source agent with smart model routing, BYOK, fully self-hosted.

CLI MCP package manager & registry for all platforms and all clients. Search & configure MCP servers. Advanced Router & Profile features.

Use Code Llama with Visual Studio Code and the Continue extension. A local LLM alternative to GitHub Copilot.

The full featured self-hosted Codex gateway, for teams, agents and you

ToolHive is an application that allows you to install, manage and run MCP servers and connect them to AI agents

Use OpenCode Zen's free models in any coding agent. OpenAI + Anthropic compatible, so Claude Code works too. No account, no API key.

Use your ZCode account in Claude Code, Codex, Aider and other coding assistants through a local proxy.

Cross platform GLM proxy for AutoClaw's Zhipu AI models. Provides OpenAI-compatible Chat Completions and Anthropic Messages APIs for Claude Code, Cursor, Continue, OpenCode, and SDKs

Use Codestral Mamba with Visual Studio Code and the Continue extension. A local LLM alternative to GitHub Copilot.

Codellama Instruct OpenAi style api.

Open Source AI with Granite and Granite Code

world-model-mcp is a signed audit memory server for AI coding agents, delivered as an MCP tool. Every event Ed25519-signed and Merkle-chained; offline-verifiable against pinned public keys. Hosted platform with hybrid post-quantum signing at etch.systems.

CustomLLM config to leverage watsonx LLMs with continue.dev.

A fully local multi-agent AI board — CEO, CTO, CFO, CPO, COO running on your own GPU/CPU. No cloud. Just Local Models.

Local guardrails for AI coding agents. Checks each tool call from Claude Code, Cursor, Codex and MCP servers, and stops or asks before destructive ones run (DROP TABLE, rm -rf, force-push, unscoped UPDATE/DELETE). Free, runs on your machine, source-available under the Elastic License 2.0.

MCP server giving AI coding agents offline, token-efficient persistent memory — recall decisions, patterns, and bugs across sessions with 68.5% fewer tokens than re-reading context. No LLM, no server; works with 15+ agents.

Model Context Protocol server giving AI assistants live access to moodledev.io docs, Hooks API, capabilities, XMLDB, and the Moodle Jira tracker.

Universal AI API proxy — hot-switch between any LLM provider (GitHub Copilot, OpenAI, Anthropic, Z.AI, MiniMax, DeepSeek, Groq, Ollama, and more) from any AI coding agent (Claude Code, ForgeCode, OpenAI Codex CLI, OpenCode, Cline, Aider, Continue)

Drive Unreal Engine 5.7 from any MCP client: 155 editor-automation tools (inspect, author, render, DMX) over a local TCP socket.