Tool Use Python Packages

tooluniverse

Democratizing AI scientists with ToolUniverse

geny-executor

Manifest-driven 21-stage agent pipeline for Python — 5 LLM backends (Anthropic / OpenAI / Google / vLLM / Claude Code CLI), tools, skills, memory, sandboxed CLI runs, MCP. The engine behind Geny. Apache-2.0.

agentcrew-ai

Chat application with multi-agents system supports multi-models and MCP

litestar-mcp

Litestar plugin for Model Context Protocol (MCP) — expose route handlers as MCP tools and resources for AI agents.

autoagent-core

Zero-dependency LLM agent runtime — tool use, MCP, sandboxed self-written tools, checkpoint/resume, approval gates, bi-temporal fact memory, code-level guards (context bounds, loop guard, egress control, shadow mode), declarative tool policy, OpenTelemetry. Not a framework. 885 tests, CI Linux/Windows, 31 runnable demos, online visual builder.

evalarc

Review model and agent regressions behind better scores. Compare every changed check, inspect original model outputs and tool traces, and reproduce the decision offline.

pawnlogic

A modular AI Agent framework designed for complex task orchestration and logic reasoning.

api-mcp-compiler

Compile REST/OpenAPI and SOAP/WSDL services into governed MCP tool surfaces: provenance on every field, policy derived separately from code, and a human approval gate before any write or destructive tool is emitted.

agentu

A harness-engineered AI agent runtime. Build agents with tool isolation, self-correction, and permission scoping out of the box.

toolregistry

ToolRegistry: A Protocol-Agnostic Tool Management Library for Function-Calling LLMs (OpenAI, Anthropic, Gemini, LangChain, MCP)

siglume-api-sdk

SDK for publishing APIs that AI agents subscribe to on the Siglume API Store. Python + TypeScript. Earn 93.4% of subscription revenue.

agentdog

AgentDog helps developers inspect, test, score, and monitor AI agent runs locally.

bigcodebench

[ICLR'25] BigCodeBench: Benchmarking Code Generation Towards AGI

agenticx

AgenticX is a unified, production-ready multi-agent platform — Python SDK + CLI (agx) + Studio server + Machi desktop app. Features Meta-Agent orchestration, 15+ LLM providers, MCP Hub, hierarchical memory, avatar & group chat, skill ecosystem, safety sandbox, and IM gateway (Feishu/WeChat).

kani

kani (カニ) is a highly hackable microframework for tool-calling language models. (NLP-OSS @ EMNLP 2023)

fabri

Local memory + orchestration layer for custom LLM agents: SCOPE-style tactical/strategic memory over execution traces, Qdrant-backed, polyglot subprocess tools, agent-as-tool composition.

dendrux

Async Python runtime for agents that survive failure, persist evidence, and pause for the real world.

mcp-placard

Static analysis and drift detection for MCP servers: enumerate the surface a server exposes to an agent, classify every tool's blast radius (R0-R5), hash it, and fail CI when it escalates. Never invokes a tool.

verel

Verified agents — nothing is "done" until a grader returns a verdict. Eyes (AgentVision) + verdict bus + compounding memory + a fleet + agent-built tooling + agent-run CI/CD.

intaris

Guardrails service for AI agents. Default-deny tool call evaluation with LLM safety analysis, priority-ordered decision matrix, and human-in-the-loop escalations. Session recording, behavioral analysis, MCP proxy, secret redaction, and real-time audit.

tool-scorer

Python framework for evaluating LLM tool-calling behavior with comprehensive metrics on accuracy, efficiency, and correctness

miniverl

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

agent-action-guard

🛡️ Safe AI Agents through Action Classifier

document-adapter

LLM이 DOCX/PPTX/HWPX 문서를 직접 편집할 수 있게 해주는 통합 어댑터 + MCP 서버. Claude Desktop / Claude Code / Anthropic API Tool Use 호환. pip install document-adapter