Python agents

Unified CLI for the six-tool agent stack (Node.js tool, installable via pip)

Dead-simple Groq LLM chaining in Python. Chain prompts with .step() — no LangChain needed.

Native LangChain chat models for OpenAI Codex subscription OAuth and OpenCode Zen/Go — no Node sidecar.

LLM client layer for the Lexigram Framework — OpenAI, Anthropic, Ollama, Cohere, Groq, Mistral

LlamaIndex tool spec for the Ejentum Reasoning Harness. Wraps the hosted Ejentum MCP server and exposes eight cognitive harness tools as LlamaIndex FunctionTool objects: four dynamic (reasoning, code, anti-deception, memory) plus four adaptive variants (adaptive-reasoning, adaptive-code, adaptive-anti-deception, adaptive-memory) that pre-fit the cognitive operation to the caller's task via an adapter LLM.

Route prompts to the right LLM by complexity. Supports Groq, OpenAI, Anthropic. Save cost, keep quality.

Lightweight evaluation framework that unifies inference through a single VLLM sampler and runs IF-EVAL, IFBench, WritingBench, HealthBench, Arena-Hard, and AlpacaEval end-to-end.

Local-first, CI-friendly regression testing for LLM outputs.

Decorator-based LLM-native typed Python functions — declare the signature and docstring; the LLM runs the body.

Static preflight scanner for MCP servers and agent skills

Public-safe MVR API preflight wrapper for AI agents.

loading...

Standalone observed-fact graph substrate for code and document structure

Point your agent at your docs and your RAG app; get a golden test set + an LLM-as-judge & retrieval scorecard, in one command.

Accountability infrastructure for AI agents

Enterprise-grade Python RAG stack — loaders, cleaning, chunking, embedding, and vector store in one composable library

Production-ready RAG with near-zero hallucination. Pluggable providers (OpenAI, Anthropic, Gemini, OpenRouter, Azure, Voyage, ...), graph + agentic + multilingual strategies, a FastAPI server with a bundled React UI, and one-click publishing of a configured setup as a public ask-only assistant.

TraceLLM — Open-source LLM observability, tracing, and replay infrastructure.

Document-grounded Q&A with sentence-level citations and faithfulness verification

A minimal terminal screensaver for when AI agents are working

Circuit breaker SDK for LLM agents.

AgentCamp SDK — where agents train their own experts. Observe, train, shadow, and activate cheaper expert models behind your existing agents.

An agentic middleware for automated JSON validation and self-correction from LLM outputs.

An MCP which adds terminal capabilities to an agent.

Official Python SDK for AgenticDome security guardrails, DLP, tool authorization, and multi-agent delegation enforcement.

Universal FastAPI and FastMCP gateway backed by a local Pi AgentSession

A minimal, reusable Python framework for building LLM agents with tools and an observe-act loop.

Agent orchestrator for terminal-based coding agents — discovers installed CLIs, runs them in parallel, ships giant projects in hours

Graph control for AI-agent workflows.

AI agent and MCP server security scanner — discovery, static analysis, supply chain audit, and multi-agent trust analysis

Stateful coordination layer for multi-agent AI systems

Autonomous coding copilot

Console logging utilities for Autourgos agents and orchestrators.

Optimization config loader for Azure AI Hosted Agents

Store, retrieve, and transform AI agent memories backed by Azure Cosmos DB

Multi-tenant agent memory framework powered by BECOMER — zero tokens, any LLM

An always-present chat-agent sidebar in the Django admin, powered by django-ag-ui.

Production-grade read/write split RAG retrieval system

Eight-layer middleware guardrail pipeline for LLM-powered personal finance applications

Wrap any AI agent in a rootless Podman, Docker, or Kubernetes container for automatic isolation.

Normalize LangChain, MCP, and multimodal content blocks into provider-ready text and image payloads.

Provider-agnostic persuasion guard built with LangChain and LangGraph.

LangChain-based read-only query layer for LightRAG knowledge graph

Local auth surface manager for LLM API keys and OAuth tokens.

Add your description here

A lightweight LLM inference engine supporting OpenAI-compatible APIs.

Eval gates with confidence intervals: deterministic gates, calibrated LLM-as-judge, agentic-trace evals, and a statistically honest regression gate for LLM pipelines

A control layer for LLM-in-the-loop reliability: wrap an unreliable generator in a verify-and-refuse feedback loop with a pluggable controller (deterministic checks, a model critic, or compositions) and composable 'circuits'. Keep one exact element in the loop and outputs are checked, corrected, or honestly refused. Zero runtime dependencies; runs with no model at all.

Declarative slot-based onboarding schemas for LLM-driven conversations.

Universal Python LLM Provider Abstraction Library

Real-time music generation models.

Python SDK for mink — sandboxed AI coding agent

Redis Agent Memory integration for NVIDIA NeMo Agent Toolkit

Production helpers for running Ollama under concurrent load across local and remote endpoints.

OpenAI-compatible client + worker that bridge inference requests over a Redis queue (Redis Streams), for running LLMs across heavily restricted networks.

AgenticDome security middleware and firewall plugin for OpenClaw.

Connect your OpenAI agents to Oversee — two lines of code for full observability

Python bindings for ovstorage

Image tiles to vectors to FAISS index. Three independent CLI tools.

Build searchable FAISS indexes from any document source

Document to image tiles. Renders web pages, PDFs, and local files as tiled screenshots.

FAISS-based visual search API. Serves any pre-built index.

Persistent, concurrent-safe project memory for Claude Code agents

Knowledge flywheel MCP server + academic paper research pipeline for Claude Code and VS Code Copilot

MongoDB Atlas-backed persistent memory capability for Pydantic AI / pydantic-ai-harness.

pytest-compatible test harness for AI agents — deterministic record & replay for Anthropic Claude

A pluggable, self-correcting agentic RAG framework with MCP-native retrieval.

Runtime hallucination detection middleware for RAG applications

The Agent That Kills All Agents — Self-hosted local AI agent

StorageOps — S3-compatible object storage diagnostic toolkit. A Pi Coding Agent extension + skill pack.

Shared runtime for Superpos slim agents (Telegram bot, Superpos poller, worktree manager, telegram streamer).

Three minds, one context — Multi-agent AI orchestrator for Claude Code, Codex, and Antigravity CLI.

AI-powered UI element detection and device automation framework

OpenTelemetry-based trace SDK for Dockerized Python Agents.

AAEP integration for LangChain agents via callback handler

Local-first, MCP-native multi-agent platform — job tracker + daily briefing for software engineers, with tracing, evals, and budgets.

LangChain callback adapter for AeroGraph — automatically record LangChain traces.

Visual debugger and observability dashboard for PydanticAI agent pipelines

MongoDB Atlas-backed thread/session persistence + semantic/episodic vector memory for VRSEN Agency Swarm.

Thin Python SDK surface for the stateless Agent Control Specification runtime

Agent x402 Paywall MCP — wraps Coinbase x402 (HTTP 402) so agents pay per-call without Stripe accounts. On-chain settlement via USDC (Base/Polygon/Solana) or Lightning. The plumbing for the agent economy. By MEOK AI Labs.

Open-source reliability testing for tool-using AI agents: record, profile cost, detect regressions before production.

Run automated tests against Salesforce Agentforce agents (External + Internal Copilot) and score them into evidence — fully local, privacy-first, no API key required.

An open-source agentic coding harness for building and studying AI coding agents

Constitutional Governance Kernel for AI Agents — trust scoring, approvals, audit trail

Python SDK for AgentInsight — LLM observability, tracing, evaluation, and prompt management

Own your agent as a neutral folder; audit its portability, compile it to provider formats (Anthropic YAML / AWS Bedrock AgentCore / Google ADK), and deploy it live (Anthropic, AWS Bedrock AgentCore Harness, Google Agent Engine). Own the definition, rent the runtime.

Universal Agent Commerce Layer — AI agents search products across 40+ stores in 30+ countries, 52 languages. x402 micropayments, affiliate commissions, agent referral propagation.

agentwright: typed, composable primitives for building, validating, and running AI agents.

数字公民平台 — 创建、管理、社交 AI Agent 身份系统

SparkNotes for your AI agents — log runs to AgentNotes with one pip install and env vars.

Apohara ContextForge plugin for vLLM V1 — multi-agent KV-cache coordination, JCR Safety Gate (INV-15), RotateKV INT4 hooks, on AMD Instinct MI300X.

Simple agent harness made using Pydantic AI.

Custom AutoGen ChatCompletionClient adapter using the AWS Bedrock Converse API