Business agents

High-throughput scheduler & resilience toolkit for Azure OpenAI endpoints (queueing, rate-limit smoothing, backoff, tracing).

High-throughput parallel LLM agent execution with tool deduplication, structured output, and self-hosted inference support.

Unified LLM framework — 8 providers, reasoning models (thinking_budget), VLM-OCR, GraphRAG, ORPO fine-tuning, 80% test coverage

LangChain middleware for routing sensitive tool calls through Beav3r approvals.

Client library for the anthropic API with the AWS Bedrock endpoint.

Python client for AWS Bedrock Gateway with IAM authentication

Small, opinionated AWS Bedrock client wrapper: adaptive throttle, cache-aware cost tracking, and structured-output parse-and-repair. Single-cloud, single-purpose.

Audit and fix Anthropic prompt caching on AWS Bedrock through any abstraction stack.

Python SDK and CLI for BedrockRouter — use every AWS Bedrock model through a self-hosted proxy with friendly aliases and cost tracking

Low-level Python ergonomics for AWS Bedrock + Anthropic Claude: retries, cost ledger, streaming-error normalization.

Pythonic framework for AI agents. Build single agents or multi-agent teams with streaming events, parallel execution, and full observability.

High-performance JAX-based filtering for Dynamic Factor Stochastic Volatility (DFSV) models with functional architecture

A comprehensive Python library for managing AWS Bedrock Converse API interactions with multi-model support, intelligent retry logic, and parallel processing capabilities

BibleMate AI - Automate Your Bible Study; A headless version of BibleMate AI Agent Mode

BibleMate AI - Automate Your Bible Study; A headless version of BibleMate AI Agent Mode

Compatibility alias package for bijux-canon-agent. Preserves installs, imports, and the legacy bijux-agent PyPI package name and CLI command at the same synchronized version.

Compatibility alias package for bijux-canon-ingest. Preserves installs, imports, and the legacy bijux-rag PyPI package name and CLI command at the same synchronized version.

Bog Agents — a still-water agent framework. Patient by default, opinionated where it matters, batteries included. Build agents on top of any major LLM provider (Anthropic, OpenAI, Bedrock, Google, Mistral, Groq, DeepSeek) or run local with Ollama. 80+ middlewares, real subagents, file-system + shell + sandbox backends, MCP tooling, a typed config surface. Built on LangGraph; pass through in harmony.

Bog Agents Daemon — the patient watcher. Runs your bog-agents on cron schedules, file-change triggers, webhooks, and git push events; survives reboots, persists state, and reports back. Use it when you want an agent that wakes itself rather than waiting for you. Pass through in harmony.

SDK for scientific agents

Provider-agnostic LLM client with agentic capabilities. One interface, any provider.

Browser Bridge server and CLI for controlling a Chrome extension over WebSocket

A Python module for analyzing Bluesky posts and threads using AI models

An agent is just a for-loop. The simplest possible agent framework.

BuddyAI CLI — LLM-driven agentic framework powered by Claude

Library of modules used throughout the BugSwarm toolset

This is a reserved placeholder package to mitigate dependency confusion risks.

Auto-optimize LLM prompt caching. Save 60-90% on Claude, GPT & Gemini API costs.

Asynchronous, concurrent requests to the LLM REST APIs.

CEHR-GPT: Generating Electronic Health Records with Chronological Patient Timelines

Anthropic provider package for Celeste AI

Validate LLM responses with composable, production-ready checks

Pytest for your prompts — test, version and audit LLM outputs in pure Python

Chroma MCP server with Ollama embedding configuration

Interact with Google Chrome local storage

Temporal Defragmentation for Quant Finance — experimental regime representation learning

A unified Python library for Large Language Model (LLM) providers with real-time streaming, function calling, middleware support, automatic session tracking, dynamic model discovery, and intelligent system prompt generation.

Drop-in redaction proxy for the Anthropic API — anonymize prompts, deanonymize responses, log the redacted form for compliance.

A modern desktop client for the Anthropic Claude API with streaming support and extended context capabilities

Unofficial Anthropic Claude API for Python3.

A powerful CLI for Claude AI using AWS Bedrock - no Anthropic subscription needed

Track your Claude session + weekly usage. Never waste limits again.

Calculates Anthropic Claude cost savings from use of prompt caching

A command-line tool to extract and convert Claude conversations to markdown format efficiently.

Python SDK wrapping Claude CLI - Use Skills, Hooks, Slash Commands, Agents without official SDK

MCP server for async collaboration between two Claude users via shared project state

Community Python SDK for the Anthropic Compliance API

Multi-Agent Orchestration System for Claude

CLI tool to convert Claude conversation exports to Markdown files

A library for auto-generating JSON Schema and dispatching tool calls in applications that use the Anthropic API to interact with Claude.

AI-powered meeting scheduler using Claude and Google Calendar. Stop playing email tennis.

Beautiful HTML visualizations for Claude API Message objects

Display Claude.ai plan usage as ASCII meters in the terminal

Intelligent model selection for optimal cost-effectiveness with Anthropic's Claude AI

Real-time Claude Pro/Max usage monitor — tray icon, overlay widget, notifications, mini mode, sparklines

Claude Proxy Router - 智能代理路由器

Thin OpenAI-compatible API proxy backed by Anthropic Vertex AI

A pip-installable Python CLI that pipes stdin into Claude with Pydantic-validated structured outputs.

Check your Claude plan usage limits from the terminal — no more switching to the desktop app

High-performance CLI tool to find and repair broken Claude conversation sessions

Claude prompt router that chooses the right model tier and scaffold using local embeddings.

Reusable Claude Agent SDK wrapper with bug workarounds, retry, and timeout

The easiest way to build AI agents with Claude. CLI, SDK, and Cloud — 5 lines to start.

Anthropic Claude API 配置管理工具，支持多环境切换

Instantly check if your Claude usage is 2x or 1x during Anthropic's March 2026 promotion.

Automatic token usage and cost tracking for the Anthropic SDK

macOS menu bar app for Anthropic API usage and cost tracking

Async Python wrapper for the Claude.ai web app

A Python package provides an unofficial api for interacting with the Claude from Anthropic

MCP server for Claudepedia - a shared knowledge base for Claude instances

A command-line interface for Claude AI

Claudetools is a Python library that enables function calling with the Claude 3 family of language models from Anthropic.

Pydantic-like XML parser for Claude's XML output

Lightweight composable multi-agent framework with StateGraph, RAG, Memory, MCP, and structured output

CloudAgent Python SDK and CLI — Anthropic Managed Agents compatible platform

A model-agnostic Python library with a unified API for multiple LLM providers

An agent is just a for-loop. The simplest possible agent framework.

AI co-pilot for ComfyUI — inspect, discover, modify, and execute workflows through natural conversation

Use Composio to get an array of tools with your Anthropic LLMs.

A bureau of Claude agents that deliberates, decides, and delivers.

Tamper-proof memory, chain-of-custody, and OS-grade runtime for AI agents

ConsistentRAG: Improving factual consistency in RAG through knowledge graph grounding and multi-agent refinement

Production-Grade Agent Framework with Crash Safety Guarantees

Copass agent primitives for the Anthropic Managed Agents API

FastAPI endpoints for CopilotKit LangGraph history persistence. Companion to the npm package copilotkit-langgraph-history.

Fan-Out/Fan-In agentic orchestration framework with multi-provider LLM support and MCP tool integration.

Tools for coverage threshold limits

CrewAI tools for The Colony (thecolony.cc) — let your AI agent crews interact with the AI agent internet

The fastest way to build, chain, and reuse LLM agents and flows

Python app using llm via MCP for modeling and solving a csp problem in pycsp3

CUFEL-Q Arena 平台智能体开发 SDK

RAG indexing and search for Cursor IDE via MCP

A lightweight framework for running generative agents through a Streamlit interface.

Dagster integration with Anthropic

Simple, unified interface for all major LLMs

ChromaDB virtual filesystem backend for deepagents — instant session creation, zero marginal compute cost.

Multi-LLM agent framework with mem0-backed memory, llama-index RAG, MCP tool support, and reflection.

Scalable multi-provider LLM library with unified interface for OpenAI, Gemini, Anthropic, Deepseek, Moonshot and xAI