PyPI agents

The `bag` CLI to scaffold and deploy a bnbagent-sdk seller agent on BNB Chain.

Agent-runtime library for bnbagent-studio: wallet, ERC-8004, ERC-8183, x402 over bnbagent-sdk.

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

Record & replay the claude-agent-sdk wire for deterministic, offline tests.

Thin wrapper for `claude --resume` that dumps session history to less before launch.

Watch your Claude Code subagents work — a live office for every conversation.

Claude-to-Antigravity MCP Bridge.

Node daemon for AceDataCloud Coding Bridge — run Claude Code on your own machine and drive it from the web.

LedgerProof adapter for the Cohere Python SDK. Side-channel transparency receipts for EU AI Act Article 50.

Thread-based agents with typed streaming, tool-hide rules, branch compression, and optional Django persistence

Python SDK for the Durable Agents public API.

Make Google Gemini structured output actually validate against your Pydantic models — fixes the anyOf/enum drop, ignored numeric & length bounds, degraded array tails, and MAX_TOKENS string-runaway collapse.

Shared Google Gemini API gateway with multi-key rate limiting and retries.

A workflow-driven AI coding agent. Slash-style commands (work-issue, fix-bug, ...) are defined as markdown workflows in .ai/ and the repo's own instruction files are the source of truth.

Google-OAuth-gated LLM gateway: verify a Google ID token, then run a Gemini (Vertex AI) completion for the verified caller.

Shared FastAPI surface for the H Agent Platform agents API

LedgerProof adapter for Haystack 2.x — side-channel cryptographic receipts for EU AI Act Article 50 transparency.

Claude Agent SDK integration for Hindsight - persistent memory tools and hooks for Claude agents

JupyterHub LiteLLM API key management handler

Deterministic LangGraph orchestration with optional Codex execution

Pydantic AI integration for LatchGate — execution security kernel for AI agents

Llimona addon providing an Azure OpenAI provider integration.

Local-first macOS security audit, monitoring, and investigation agent.

Analyze test strategy across SaaS products with coverage metrics

LedgerProof adapter for the Mistral Python SDK. Side-channel transparency receipts for EU AI Act Article 50.

The self-improving AI agent — creates skills from experience, improves them during use, and runs anywhere

Natural language to skillm DSL

Multi-credential LLM client with automatic failover and provider abstraction

Production-grade autonomous AI agent framework with 1,371 tools across 100 classes and a 5-role LLM pipeline — built on the NPMAI ECOSYSTEM.

Allows to use server environment with fs storage attachment

Allows to use server environment with fs storage

Search and filter Ollama models by size, context, input type and more.

MCP server that bridges to any OpenAI-compatible chat API (DeepSeek, OpenAI, Azure, local llama.cpp, etc.).

Cost tracking and budget enforcement for Azure OpenAI API calls

Prism — the economic network for autonomous AI agents. Stripe for agents.

AI-powered project delivery MCP server for Partner Core

A production-grade Python framework implementing reusable multi-agent design patterns, stateful orchestration, and resilient execution harnesses for autonomous LLM systems.

CI security regression testing for RAG apps and AI agents

SentraGuard framework-neutral guardrail core for AI agents — LangChain/LangGraph, LiteLLM, and direct provider adapters (no OpenTelemetry dependency).

Mesh topology for distributed agent communication and workload distribution

Universal skill registry — reuse Python, Docker, CLI, REST via MCP

CrewAI tools for SNAP — private agent-to-agent payments on Solana using zero-knowledge proofs

SuperAgentX Policy Engine. Enterprise-grade AI Agent Governance Framework. Govern AI agents with policies, approvals, threat detection, cost controls, memory protection, PII, compliance protections and multi-agent security.

Swiss AI Hub Agent SDK: build transparent, workflow-based, event-driven AI agents.

TopK document store integration for Haystack with vector search, semantic search, keyword search, hybrid search and metadata retrievers

Haystack integration for transformers

Aggregate package for Turbo Agent core agentic capabilities

Voice-Interactive Reasoning Agent — a voice-controlled AI agent that accesses local folder/markdown context during collaborative sessions.

Session handover engine for AI coding agents: checkpointed, resumable, backend-agnostic persistent memory.

Isolated sandbox for AI coding agents

Harness- and transport-agnostic agent runtime: supervises an agent's process lifecycle independent of its harness (brain) and transport (IRC/Slack/agtag).

Manage Claude Code & Codex sessions, cost, and data from your terminal

Universal vision tools for AI agents via Model Context Protocol

AgentAudit SDK — make any AI agent's decisions independently verifiable on Algorand.

Local-first Maildir-inspired work mailstore for software agents

The official Python REST client for the Agentex API

Agentic QA Maestro — fully automated testing from JIRA stories to defect management

Agentic TestOps for Python projects: automated pytest execution, failure diagnosis, and repair advice.

Outcome-verified agent trajectories, benchmarks, and RL environments — with a live leaderboard and a CI gate. Offline-first.

Open-source runtime tracing and diagnostics for AI agent execution flows.

Lightweight drop-in wrapper for the OpenAI Python client that logs telemetry and detects PII risks.

Haystack components for structured KV extraction from financial PDFs via Azure Document Intelligence

Per-task budget, loop detection, and kill-switch middleware for agent LLM calls. v0, deterministic and dependency-free.

Evidence-gated SDLC for Claude Code, installed as configuration (no app code, no Docker). Asks ordered questions and installs CLAUDE.md + .claude/ (rules, the chosen profile's agents/skills, hooks, artifact templates) + optional .mcp.json; run /sdlc to drive spec → review → build → test → security → ship through profile-aware quality gates with a tamper-evident gate ledger, working memory, and a self-improving learnings loop.

CLI tool to explore Claude Code conversation history

View Claude (claude.ai) usage limits across multiple accounts from your terminal

Universal agent memory plugin (ForgetForge): recall-centric retention, tiered forgetting, Rust scoring engine for Hermes, Claude Code, and Codex.

Universal agent plugin for Cluxion preprocessing, honesty contracts, clarification, Rust work queue, and resource-aware harness handoff.

Universal agent coding harness plugin: cursor logic, safe patch, line budget, Rust index, test gates.

Agentic functional programming: a multi-agent runtime built on three primitives — recursive spawning, addressable mailboxes, and capability passing — with FP-style combinators on top.

CrewAI tool for hard spending limits on AI agents via Valta

Physical-bottleneck, reverse-consensus investment analysis for every market — an LLM agent chain that traces any stock (A-share / HK / US) to its supply-chain constraint and refuses to chase narrative-driven tops. Bring your own LLM key.

Local runtime host for long-running Deep Agents channels and schedules (experimental)

Local-first TUI coding agent for Gemma 4 + llama.cpp, with GraphRAG retrieval and a curated cookbook of domain recipes.

Generate GDScript API indexes and enforce public API comment coverage.

Google GenAI SDK wrapper for integrating the Model Context Protocol (MCP)

Google Cloud Agentidentitycredentials API client library

Agent OS Kernel — CPU-scheduling-inspired runtime for LLM agent workloads

Haystack integration for Hindsight - persistent memory for AI agents

Superagent safety integration for Hindsight - guard and redact memory operations

CrewAI tools for iFlow Search (心流搜索) — web search, image search, and web-page fetching.

Agent-first IMAP CLI for safe email search, read, attachment download, and draft creation

LlamaIndex tools for iwant.fyi -- reference implementation of the iwant.fyi demand-side protocol v1.1. Lets any LlamaIndex agent express structured purchase intent and receive matched supply across multiple sources.

Drop-in Slack-fronted Claude Code agent. Add a skills/ directory, run `jean init`, deploy to Fly.io.

Knowledge-graph-powered RAG sidecar for infrastructure querying

LangChain integration for GoodSender — transactional email API with consent management

LangChain callback handler for NeBeso AI cost tracking

LangChain integration for Neuralwatt — energy-aware AI inference.

LangChain tool for hard spending limits on AI agents via Valta

LangChain VectorStore integration for Vector Panda

A LangChain integration tool that provides reliable web scraping capabilities at any scale using ZenRows' Universal Scraper API, extended with Adaptive Stealth Mode

LlamaIndex callback handler for NeBeso AI cost tracking

Multi-model coding agent powered by NVIDIA Llama (3B, 70B) with CLI, web search, and tool-use loop

Regression canary for LLM apps: YAML test suites, offline-first assertions, baseline diffing, and agent-trace policy gates for CI

Framework open-source para avaliação de confiabilidade de chatbots baseados em LLMs

Building Language Models from Scratch — tokenizer, Transformer, training, inference, and fine-tuning in clean, tested PyTorch.

Cache-optimized LLM conversation history management with static/dynamic system prompts, transition modes, and compaction hooks.

Policy-gated, durable, audited execution for AI agent tool calls. Every action gets a five-verdict policy check, a crash-safe checkpoint, and a tamper-evident audit event — a library, not a server.

This module adds support for serving a /llms.txt file in the website root. The content can be configured per website in the website settings.

An MCP server that onboards you to any codebase — point it at a repo and ask questions like a senior engineer who knows the project inside out.