#llm

Utility-first helpers for the Web Streams API, aimed at streaming and transform pipelines.

no! more! slop!

Official Go SDK for the Webclaw web extraction API

该仓库未发布描述。

Official MCP server for OrcaRouter — the LLM gateway with auto-routing and fallback chains. Browse models and chat from Claude Desktop, Cursor, Windsurf, or any MCP client.

LLM-Friendly Convention Linter for AI Coding Tools

Python-side witness stack for the official DeepSeek Harness (Node). Five `dsh doctor --node` probes each report something the Node runtime will not. `pip install deepseek-harness-cli` first

Karma is a powerful, modular library designed to simplify common tasks in Go development. Whether you're handling authentication, parsing SQL, managing middleware, integrating with third-party APIs, or dealing with file management, Karma has got you covered.

Rust based JSON repair library

MCP bridge that drives four coding CLIs — Google's Antigravity (agy), OpenAI Codex, GitHub Copilot, and Cursor — as sub-agents inside Claude Code. Text answers, image generation, real repo work, and parallel swarms, on quota you already pay for.

ローカルLLM（mlx / mlx-vlm / llama.cpp / router）を OpenAI 互換 API として起動・管理する軽量サーバー

Build programmatic SEO pages that stay consistent.

Multi-model code review council — consensus-driven PR reviews

Llama cpp + CapacitorJS support

Provider-agnostic Go AI framework for building RAG, agents, and AI-powered applications. Swap LLM providers without code changes — supports Ollama, OpenAI, and more.

Enhanced fork of OpenWolf — a token-conscious second brain for Claude Code, with bounded storage, self-maintenance (openwolf doctor), .wolfignore scoping, and tunable retention. AGPL-3.0.

TypeScript SDK for Boardwalk workflows

Claude Code (Anthropic) provider for the intisy-ai AI-proxy ecosystem.

🎭♪ Figaro — a coding agent. Pronto prontissimo.

A small, nestable agent-loop job primitive: loop(body→until→review) composed with dag, pluggable engines, agent-judge gates, budget + run records. TypeScript.

Cut ~90% of Claude Code's context usage while keeping nearly all the memory

Multi-perspective US stock ticker analysis — fans out 4 expert personas (Value/Growth/Quant/Options) × 3 timeframes (Long/Mid/Short) with Zod-validated LLM analyses via @krdn/llm-gateway

One workflow framework, every coding agent. Install your AI-coding discipline into Claude Code, Cursor, GitHub Copilot, Gemini CLI, Codex, and Windsurf from one tool-agnostic core. Apache 2.0.

The governance layer LLM proxies promised as a TypeScript library you can read. Every call runs through one governed pipeline: rate limit → spend caps (per-user + global circuit breaker) → cache → schema-validated provider failover → usage log → LLM-judge No sidecar proxy to deploy.

Zero-dep streaming markdown for the browser — Rust→WASM, 100% CommonMark 0.31 + GFM. Built to not melt the main thread on concurrent LLM streams.

Skillscript — a small declarative language for authoring agent workflows. Runtime, compiler, and CLI.

Fnord is an AI code archaeologist.

High-performance enterprise-grade large model gateway for the power ecology of large models (LLM), equipped with rich traffic governance strategies.

Create agents that helps your customers right into your apps

A Model Context Protocol (MCP) server implementation that integrates with WebScraping.AI for web data extraction capabilities.