Source
PyPI agents
25,155 AI agents indexed on MeshKore from PyPI. Agent frameworks and tools published to the Python Package Index. Each entry links back to its project page.
25,155 agents · ranked by popularity · refine in the directory →
Source platform: PyPI →
PyPI agents — page 220 of 252
CI evaluation framework for RAG and AI agents: groundedness, retrieval quality, hallucination, citations, latency, cost, and regression gating before production.
Configure ragas RAG evaluation to use AINative's free LLMs + embeddings — no OpenAI key needed.
Respan instrumentation plugin for Groq
Respan instrumentation plugin for Mistral AI
Respan instrumentation plugin for Ollama
Salesforce REST/SOQL/Bulk API 2.0 connector + MCP Server + A2A Server for Agentic AI!
Haystack integration for SerperDev Web Search
Alias for the 'shunt' package — routes LLM requests between cheap and premium models
Distributed RAG QA-evaluation dataset generator for local files and Confluence.
Haystack integration for spacy
Fast C-backed trigram language model for word prediction and sentence completion
AgentLedger Python SDK — track, attribute, and govern AI agent spend across OpenAI, Anthropic, and Gemini.
LangChain VectorStore integration for VectorDB
Multi-agent framework for generating Rivet analysis routines using large language models
Authorization for AI agents: verifiable, scoped, revocable capability tokens (mandates) with attenuable delegation, for MCP and A2A. The reference implementation of agent authority — zero-dependency TypeScript & Python.
A local, filesystem-backed message board that lets AI coding agents open private channels, exchange messages and artifacts, and tear them down when done.
A runtime and definition-time security guardrail framework for AI agents and developers.
Lightweight Loop Engineering orchestrator for self-hosted agent loops
Governance proxy for every AI tool — PII/PHI masking, prompt injection detection, anomaly alerts, semantic cache, token quotas, vendor routing, compliance reports
Nexus Protocol SDK — Ed25519 DID identity for AI agents. 15 lines to register, post, and vote.
Drop agents, not code — zero-code multi-agent API platform framework
AI Agent Security Scanner — detect prompt injection, tool auth bypass, data leaks
Verigent trust verification for Microsoft AutoGen multi-agent systems
Bobo — a personal AI agent for your terminal, Obsidian, and email
Capsule Gate for agentgateway (name claim; implementation to follow).
claude-autosar - AUTOSAR BSW 双格式 (.arxml + .xdm) 配置 IO / 模板 diff / lint / verify 工具集
Persistent memory for Claude Code — checkpoint-based replanning across sessions
Verigent trust verification for the Claude Agent SDK
Claude Code hook dispatcher — forwards lifecycle events to Claude Monitor
LLM usage telemetry SDK for Cloptima reporting, attribution, and analytics
CrewAI tools for the OptionsAhoy equity-compensation calculators.
Verigent trust verification for CrewAI agent delegation
Doraemon Cyber Team — Multi-server Ollama agent
High-performance unified Docker and Podman container management library
Run a prompt through a LangChain (system + human) chain over Gemini on Vertex AI; optional LangSmith tracing.
Verigent trust verification for LangChain agents, chains, and tools
LlamaIndex memory integration for Recollect
LlamaIndex tools for the OptionsAhoy equity-compensation calculators.
Provider-agnostic, reproducible, typed LLM-as-a-judge — a small primitive you can depend on.
Validate API keys for OpenAI, Anthropic, Google Gemini, Groq, Mistral, OpenRouter, Cohere, DeepSeek, Together, Fireworks, xAI, and Perplexity. Zero dependencies. CLI + library.
Static, deterministic LLM prompt/payload compression that cuts input tokens 30-90% with zero extra model calls.
Python SDK for Loupe — observability for LLM apps (latency, tokens, cost, errors).
MEOK AGENTS.md Linter MCP — validates the cross-vendor coding-agent spec (Cursor / Claude Code / Cline / Windsurf / Aider / OpenAI Codex) stewarded by AAIF. Security smells + consistency lint + passing templates. By MEOK AI Labs.
UK lorry-loader (hiab) compliance — ALLMI ACoP, BS 7121-4 lift planning, LOLER Thorough Examination, CPCS A36 / A40 / A73, brick-grab POD, contract-lift triage. For builders' merchants + specialist hiab operators. By MEOK AI Labs.
An OpenTelemetry-based SDK for monitoring GenAI applications.
Native LangChain chat models for the Claude Code (Anthropic) and OpenAI Codex subscription OAuth sessions, plus OpenCode Zen/Go — no Node sidecar.
Open SDK for agents that inhabit persistent game worlds.
LangChain tools for the OptionsAhoy equity-compensation calculators.
Python client for the ProxyLLM OpenAI-compatible gateway.
pytest plugin that runs your LLM app under controlled chaos (timeouts, brownouts, rate limits, MCP errors, composed failures) and asserts it still meets its contract
AI pull-request reviewer: hybrid RAG + a code graph + Claude Code. Whole-repo context, inline GitHub comments grounded on exact code.
Run SSH commands and copy files from Python — one function, no Paramiko boilerplate.
Run Claude Code or Codex on your machine against your Spryng board (the Spryng Local Agent / claude-code-runner).
SuperMemory: MCP-first learning memory layer for Claude, Cursor, and agent workflows
SuperMemory: MCP-first learning memory layer for Claude, Cursor, and agent workflows
Web Over Ollama (and Llamas) — an MCP + OpenAI router for AI desktops.
The embeddable woollama model-management core (Rust) — start of the v1.0 port.
Create redacted, shareable debug packets from failed AI agent runs.
Generate reviewable safety evidence for AI-agent-generated pull requests.
Run Claude Code as an A2A protocol agent server.
AACP coordination layer for AutoGen multi-agent workflows
Verify the merchant before your AI agent pays. A pre-pay storefront-legitimacy check (address + company-register + domain-binding + sanctions) for LangChain, CrewAI, Anthropic computer-use, or any MCP agent.
Anthropic tool_use governance middleware for Aegis ACP
LangChain governance middleware for Aegis ACP (maintenance-only — Anthropic SDK is the active hero)
OpenAI tool_calls governance middleware for Aegis ACP (maintenance-only — Anthropic SDK is the active hero)
Curriculum learning for agent training — difficulty-scored stages
Adversarial scenario generation and testing for coding agents
A terminal-first CLI for LinkedIn feed, profiles, search, and core social actions
Install, scaffold, run, and chat with your Agentberg trading agent.
AgentDrive
Portable, local-first source of truth for your AI coding-agent setup. Placeholder release; in active development.
An MCP which adds terminal capabilities to an agent.
The physical-mail API for AI agents. Send a real, tracked, compliant letter with one call.
Agentopolis — Habbo-style visualization of Claude Code agents building your repo as a city
Public chat for AI agents - WebSocket + JSON, 3-method SDK, no UI required
Microsoft Azure Napsteromniagentapi Management Client Library for Python
A pure Python terminal code agent — pluggable LLM providers, fine-grained permissions, multi-agent coordination, and an extensible skill system.
Defensive package-name reservation notice for a potential supply-chain risk.
Desktop and LAN web dashboard for Claude Code hook events
Portable multi-agent ultracode orchestration with a 3-agent adversarial debate that converges to unanimous consensus
Privacy-preserving CLI AI agent for querying local code repositories using RAG
RAG MCP Server — watches folders, chunks documents, serves semantic search via MCP
DPX settlement and intelligence tools for CrewAI agents — quote, ESG scoring, oracle status, settlement, and macro intelligence.
Python library for extracting structured event data from text posts with LLM agents.
Explain.io FinOps Airflow Cost Agent
Profile and classify agent behavior patterns from traces
Persistent agent daemon — systemd for AI agents
Skill registry, decomposition, and LoRA building for coding agents
Orchestrate micro-agent swarms using Markov transition matrices as handoff patterns
Datadog for AI agents — real-time observability, token tracking, cost estimation, error rates
The open-source, self-verified coding agent. Generate → Execute → Verify → Recover.
HumanEval for tool use — a standardized benchmark for evaluating LLM tool-use capabilities
In-memory storage backend for fast-feature (zero dependencies).
Practical, ready-to-use tools for LLM generated XML
FP - 基于生命周期钩子的插件化 Agent 框架
Reserved for Garrison, Agile Defense's secure agentic AI platform. Placeholder release.
Reserved for Garrison, Agile Defense's secure agentic AI platform. Placeholder release.
Vaultak runtime security components for Haystack pipelines