Zed-first ACP v1 server driving DeepSeek Harness (DSH): standalone stdio bridge with token streaming, tool calls, approval, plan, and session management

DeepSeek LLM provider for @namzu/sdk. Chat Completions with streaming, tool use, and thinking mode mapped onto the kernel's reasoning blocks.

Google Drive streaming tool.

A production-oriented llama.cpp LLM provider for DeepSeek Harness, with first-class Qwen reasoning support: streaming tool calls, adaptive inference policies, multi-endpoint reliability, capability-aware routing, discovery, observability, and diagnostics.

Pure logic toolkit for agent streaming, tool orchestration, and argument accumulation.

Unified TypeScript SDK for OpenAI, Anthropic Claude, Gemini, Mistral, DeepSeek, xAI Grok, Qwen and Llama — one API with streaming, tool calling, structured output, guardrails, memory and multi-agent orchestration.

AragonAgent core engine — LLM provider adapters, streaming, tool registry/execution, steering, idle watchdog, and an optional isolated-vm sandbox.

Claude for @sigx/ai — a LanguageModel on the official @anthropic-ai/sdk: streaming, tool use, adaptive thinking, refusal handling

Zero-framework LLM router for Node.js — unified streaming, tool calling, and usage tracking across OpenAI, Anthropic, Google, Ollama, and any OpenAI-compatible provider.

`@xpert-ai/plugin-anthropic` provides a model adapter for connecting Anthropic's Claude models to the [XpertAI](https://github.com/xpert-ai/xpert) platform. The plugin enables agents to invoke Claude series models with support for streaming, tool calling,

A universal LLM client with transparent provider failover, streaming tool execution, pluggable reasoning, and native observability.

Unified LLM API with automatic model discovery and provider configuration

VoltX AI — Unified LLM provider abstraction with streaming, tool calling, and structured output

Third-party, complete DeepSeek Harness agent over the Agent Client Protocol (ACP) — a superset of the official automation-only @deepseek-ai/dsh-acp: streaming, tool calls, sessions, slash commands, agent presets, config options. npx one-command, offline a

OpenAI-compatible ModelAdapter for @rejelly/core: streaming, tool calling, structured outputs, multimodal

Self-hosted LLM gateway and SDK for Node — a LiteLLM-style unified API for 13 providers (Anthropic, OpenAI, Gemini, Mistral, Groq, xAI, DeepSeek, OpenRouter, …) with per-call USD cost tracking, streaming, tool calls, vision, speech-to-text, and built-in O

A TypeScript port of the codex-rs agent core — runs the Codex sampling loop (Responses API streaming + tool calls + auto-compaction) directly in Node or the browser, with no spawned process.

Zero-dependency TypeScript client for OpenAI, Anthropic, Gemini, Ollama, LM Studio, Groq, OpenRouter, DeepSeek, Mistral, xAI, Together and any OpenAI-compatible API. Real streaming, tool calling, structured output (Standard Schema or JSON Schema), images,

Lily agent SDK — the same streaming, tool-using agent in the browser (WebAssembly) and on the server (Node). LLM + credits metered by your agent-server.

AI Agent SDK for Vue3 projects — chat, streaming, tool authorization, batch tasks