Python agents

Dynamo Protocol gateway plugin for LiteLLM: offline allowance-credential verification, 402 fail-closed enforcement, async unit metering

LlamaIndex reader for Hydrafetch. Load clean, LLM-ready web data into an index.

llama-index tools magic_hour integration (AI video and image generation)

llama-index tools wavespeed integration

Is your llama.cpp decode memory-bandwidth-bound? Measure it, in one command.

A programmable layer for unattended AI coding work: drive Claude Code, Codex, and Gemini CLIs from a YAML playbook with verify/repair loops, budgets, cross-agent fallback, and usage-window riding.

One vendored litellm-catalog snapshot + one public overlay + shared lookup/cost math for portfolio LLM pricing

Unified LLM abstraction with provider pattern for portfolio projects

LLM plugin for tiyuvta prepaid inference (api.tiyuvta.ai)

A pytest-native security testing framework for LLM applications (OWASP LLM Top 10).

Open-source debugger for AI agents.

MCP server for HiAgent OpenAPI

Host-neutral spatial intent types for XR integrations

NeuForge x402 Token Metering for AutoGen

NeuForge x402 Token Metering for CrewAI

NeuForge x402 Token Metering for Haystack

NeuForge x402 Token Metering for LangChain

NeuForge x402 Token Metering for LlamaIndex

Turn ad-hoc subagents into durable, accountable AI teams.

MemorySync memory for the OpenAI Agents SDK: a drop-in Session implementation with durable server-side history, long-term memory instructions, and agent memory tools.

Launch ephemeral agents for single tasks in OpenShell sandboxes.

OpenTelemetry Official Qwen-Agent instrumentation

OpenTelemetry smolagents instrumentation

A key-addressed retrieval system for coding agents

CPU-First Multimodal Document Chunking, Ingestion & Retrieval Engine using Native PDF Extraction, OCR, Multi-Vector Embeddings, and MaxSim Late Interaction

parakeet-index llms watsonx integration

Parlot turn-model instrumentation for LangGraph / LangChain (same debugger as voice)

Python client for Pepkio reagent-qr-label-maker tool

LLM-assisted Playwright Page Object Model generator with live locator verification.

Privacy-preserving middleware for embedding pipelines and vector search.

PyBedrock Net is a Python Minecraft Bedrock networking toolkit for RakNet transport, protocol packets, authentication, world data, entities, inventory, chat, commands, events, and bot automation.

Agentic verification harness for QuanLLM-v2.0-qm

Security testing and authorization-aware retrieval for production RAG systems.

Unified interface for RAG and information-extraction engines with built-in benchmarks.

Detect stale, drifted, and ghost documents in RAG vector indexes

Fast, deterministic & ONNX-powered hallucination guardrail & citation verifier for RAG

A lightweight RAG debugging toolkit: understand why your RAG application gave a bad answer.

Python client for the Sandbox Agents control-plane: agents, sandboxed sessions, streamed runs.

AI 多智能体协同安全检测系统

Custom hybrid RAG library

Multi-Cloud Attack Chain Intelligence Tool — See the full attack before the attacker does

Estimate the water footprint of LLM prompts, the way you estimate their cost.

LangChain / LangGraph adapter for sonar-tracing.

Multi-provider LLM adapter plugin for Spakky Agent

Subagent MCP gateway for Codex to orchestrate external coding agents through native harnesses.

Complete T4L Trainer agent package with embedded server and in-app chat loop.

LiteLLM ↔ TrigGuard policy gate — four-state authorize before routing

A Model Context Protocol server for TastyTrade brokerage accounts — monitor portfolios, analyze options, and trade

Exact tokenizer and chat-template frontend for TuiskoLLM

Local LiteLLM proxy for GitHub Copilot models with OpenAI and Anthropic-compatible APIs.

Turn hours of local video into timestamped, citable, LLM-ready text. Fully offline. Resumable.

Beam search plugin for vLLM v1 with CoW block table forking

Post-call evaluation for voice agents using audio and transcript signals.

Deterministic observability for AI agent trajectories and tool execution failures.

WorkSwarm

Pure B2A Autonomous Agent Real-Time Data Oracle & x402 Signed Grounding Engine on Polygon, Base, and Arbitrum

Fast, accurate agent-to-Android control over plain ADB: semantic UI addressing via uiautomator, no on-device install

Use AegisDB as a LangGraph long-term-memory store

Differential testing of budget enforcement semantics across AI agent frameworks

Unified detection, configuration, and chat discovery registry for AI coding agent CLIs

Deterministic cross-language guardrails for agent-assisted development

Run Codex CLI and OpenCode CLI simultaneously on isolated git worktrees - LLM task planner, parallel headless runners, auto-merge

The Unified, Non-Destructive Model Configuration & Snapshot Manager for AI Coding Agents.

Agent Failure Registry - a neutral namespace and crosswalk for AI agent failure modes

Pointer to the pypi package `agentic-graphrag`

ALS-RS agent harness, extension adapter, KB/MCP tools, and TE2 integration.

零依赖的 OpenAI 兼容模型智能路由调度器

Idempotent execution layer for LLM tool calls: outbox + dedupe keys.

Vendor-neutral reliability engineering for autonomous AI agents.

Watch your multi-agent AI workflow as a 2D retro pixel-art game

A multi-provider web search layer for AI agents

Fail-closed AI-agent authority verification, delegated capability enforcement, and signed decision receipts

Run coding agents in a per-project dev container sandbox.

End-to-end encrypted messaging between AI agents, with a directory and store-and-forward delivery

Open source AgentCICD Engine folder runner

Agent Deed — placeholder. Nothing to install yet.

Secure agent SDK for B2B teams.

A drop-in NSGA-II replacement whose authored machinery installs only when measurement says it beats the classical path

Provider-agnostic QA testing for LLM-based agents: accuracy, robustness, fairness, and security checks with pass/fail reporting.

A minimal-privilege proxy for MCP tool calls

Agentic GraphRAG: graph-based retrieval-augmented generation with agentic reasoning.

AgentInfoView (AIV) — a presentation surface & feedback channel for coding agents. Name reserved; project in active development.

Lightweight, security-first converter between coding-agent config formats (CLAUDE.md, AGENTS.md, Cursor rules, MCP configs, SKILL.md).

Portable agent profile definition — schema and parser

Portable, policy-friendly system function tools for AI applications.

A circuit breaker and runtime protection SDK for AI agents.

One skill-driven agent core, many domains.

Alibaba Cloud AgentCore (20260804) SDK Library for Python

ATIF-based reference implementation for portable coding-agent trajectories with explicit fidelity and loss reporting.

BYOC (Bring Your Own Cloud) — universal storage abstraction for storage your users already own

Spatial domain detection from cell-type motifs on single-cell spatial maps.

Summarize & export a Claude Code session into one clean handoff.md for Gemini, GPT, or another Claude.

A focused terminal workspace for collaborating with ACP coding agents.

Conduct Guard as a LiteLLM guardrail — enforce runtime policy on every LLM call routed through LiteLLM.

CrewAI integration for LastSearch — research infrastructure with Grounded Intelligence for AI agents

CrewAI tools for the Web Metadata, OpenGraph & Contact Extractor API — give your crew URL-to-structured-data extraction, AI-ready Markdown, contact discovery, and SEO auditing.

A retrieval substrate for AI agents: every query compiles to an explainable, replayable plan over one versioned, tenant-isolated store.

dbt-native Snowflake Cortex Agent materialization with skill and evaluation tooling

Pre-execution Decionis policy gates for current AutoGen tools and legacy UserProxyAgent.

Gate CrewAI tool calls on a signed Decionis Decision Dossier before execution.