Python agents

Visual browser QA console for AI coding agents — live browser, element selection, source mapping, RTL/a11y/perf audits, and test-fix-verify loops.

Evaluation package that allows benchmarking of agentic AIs from various sources and frameworks by producing statistical results which can be compared across different use cases and datasets.

A quantitative trading system development framework for A-share stock market

A private task queue system for agents using Redis + Celery with gRPC interface (Client Package)

Agentic RAG framework - Coming soon. Part of the AIH-Infra ecosystem.

RAG MCP server: ChromaDB + sentence-transformers, exposes ingest/search/list/delete tools.

put your agents on guardrails

Intelligent rate limit handling for AI agents

Fleet-wide shared rate limiter for A2A + multi-MCP deployments. Most MCP servers rate-limit independently — a hostile agent hitting 10 MCPs gets 10x quota. This MCP is the shared counter: every MCP checks here before allowing a call. Sliding window + concurrency grants + signed enforcement attestations. By MEOK AI Labs.

Decentralized agent reputation protocol — signal composition, portable reputation bundles, Merkle verification, and anti-Goodhart architecture

Role-Based Access Control for AI Agents with ML-Powered Anomaly Detection

A lightweight Python protocol for agent-oriented documentation

Intelligent Reader of various types of media

Benchmark how agent-ready a code repository is for LLM coding agents.

Score any URL for AI-agent readiness — llms.txt, JSON-LD, AI-bot robots.txt, canonical, MCP, meta, sitemap.

Shared schema, dataclasses, and version constants for the agent-readiness insights ecosystem.

MCP server exposing agent-readiness scan, detect, and apply tools to coding agents.

A tool to analyze codebases for AI-Agent Friendliness

Agent Realm — Multi-realm AI Agent deployment and management system.

Transform LLMs into robust problem-solving agents with advanced reasoning strategies

Persistent memory and AI briefings for coding agents — drop-in MCP server

Structured session checkpointing for AI agents — survive context limits, cost overruns, and session death

MCP server for validating signed agent execution receipts.

Python SDK for the Agent Receipts protocol

Reserved package - please use 'agent-recipes' instead

Real-world AI agent templates and recipes for PraisonAI

Local-first CLI that records every Claude Code and Codex session into a searchable SQLite database

A clean, elegant Python library for Redis-based multi-task scheduling with SortedSets and Streams.

Registri CLI — register, discover, and trust AI agents from the terminal

Registri discovery — capability search, matching, and resolution for AI agents

Registri framework — build, register, and orchestrate AI agents in minutes

Registri identity — Ed25519 keypairs, certificates, and cryptographic verification

Registri MCP server — discover and interact with AI agents from Claude Desktop, Cursor, and other MCP clients

Registry-driven LLM routing: build classifier prompts from agent descriptions, validate decisions, and dispatch.

Registri Python SDK — register, discover, and trust AI agents

Registri shared data models and interfaces for the Agent Registry

Detect regressions between baseline and current AI agent runs. Python port of @mukundakatta/agent-regression-lens.

A small runtime to trace, replay and debug multi-tool AI agents.

Python SDK for Agent Relay workflows

Local-first CLI for handing coding sessions between agents without losing context

Hierarchical memory management for AI agents with vector search and graph relationships

CLI for AI agents to work with Jupyter notebooks via the VS Code bridge

A beautiful terminal chat interface for OpenAI agents and DSPy Modules

Agent Reporter

Agent Reporting

Trust and reputation system for AI agents — rate, verify and build trust between agents

Portable reproducibility seed bundle for AI-agent steps (Python port of @p-vbordei/agent-rerun)

Install Claude Code skills, commands, and agents from GitHub to your project

A Python SDK to transform Claude response to OpenAI response format

Zero-dependency Python library for resilient LLM agent calls — exponential backoff, jitter, fallback chains, and dead letter queues.

A simple decorator for tracking and reverting agent actions

Official library for AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories

An Async Generator/Iterator Based FRP-Like Library for Python

Cross-session memory and analysis layer for the Rigged tool suite

A layered protocol and reference implementation for codifying risk in autonomous agent actions.

Enterprise policy, identity, approval, audit, observability, adapters, and realized-ROI controls for agentic workflows

A CLI to get training data from your own coding agents. Track rollouts and codebase snapshots at every turn. The data can be used for SFT, RL, and Continual Learning.

Universal interface for AI agents to control ROS robots via natural language

Python library for agent-browser CLI (Bundled)

Multi-Agent Roundtable Discussion Engine / 多 Agent 圆桌讨论引擎

Async routing library with pluggable routing strategies for multi-agent systems

Local documentation retrieval service for agent-assisted development. Indexes framework docs into typed Knowledge Units and exposes them via CLI and HTTP API.

Synchronize rules and skills across AI coding assistants (Claude Code, Cursor, Gemini, OpenCode, Codex)

Compare baseline vs current agent runs and surface regressions as structured reasons: success loss, new errors, failed tool calls, output drift, step/latency/cost bloat.

Model-agnostic AI agent orchestration platform with unified interface for OpenAI, Anthropic, Google, xAI, Mistral and more

Agent Runtime

Single-agent runtime SDK for A2A agents backed by MCP tools.

Python SDK for agent input scanning, tool-call guardrails, and audit logs

Agent Runtime Manager Server

Agent Runtimes

A governance and policy enforcement layer for AI agents and non-human identities

SPL (Safe Policy Lisp) evaluator for Agent-Safe capability tokens. 150 lines, zero deps, microseconds.

Safeguards for custom agents.

Tool environments for the Agent Safety Bench (ASB) benchmark

Battle-hardened safety wrapper for AI agents

Production-grade safety boundaries for AI agents - policies, tracing, replay, and human-in-the-loop approval

MCP server for AI agent safety — cost guards, injection scanning, decision tracing, agent identity (KYA), and signed receipts

One-line safety middleware for AI agent APIs. Prompt injection scanning, cost budgets, decision audit trails. Works with FastAPI, Flask, and any ASGI/WSGI framework.

Python SDK for the All-in-One Sandbox API, >=1.7.0

agent-sandbox E2B SDK patch — one-line switch to use Agent Sandbox via the E2B Python SDK

Agent Sandbox plugin for Harbor — run Terminal-Bench / SWE-bench / Harbor benchmarks on Agent Sandbox pools

Production-grade self-correcting AI agent platform with sandboxed execution

This a for iwt internal tools package

Minimal local Claude Code hook client for command risk scoring

Generate runnable AI agent projects from markdown specs in the agent-deployments repo.

agent-scaffolds

Risk-focused diffs for agent manifests, MCP configs, and tool permission snapshots.

Your personal AI assistant that scouts the web for you — find deals, compare prices, track products, and more.

CLI interface for AgentScout — your personal AI assistant agent

Core engine for AgentScout — a multi-purpose AI assistant agent

Byte-deterministic transcript format for AI-agent conversations (Python port of @p-vbordei/agent-scroll)

A lightweight, zero-dependency focused AI Agent Framework for Python.

AgentSearch: An open source framework and dataset for webscale local search.

In-process SDK runtime for agent-search

In-process SDK runtime for agent-search

Give your AI Agent eyes to see the entire internet. Search + Read 10+ platforms.

Security scanner for AI agent dependencies. One command to know if your AI stack is safe.

649 security tests for AI agent systems - MCP, A2A, L402, x402, UCP/ACP merchant-journey, AP2 mandate + Fireblocks x402 hardening + Visa TAP / Mastercard Agentic Tokens + denial-of-settlement finality, decision governance, AIUC-1 compliance, NIST AI 800-2 aligned