AI Agent Tools
Explore products that execute tasks and frameworks that help you build multi-step workflows, with clear boundaries, use cases, and source signals.
See the best AI agent tools →How we define agent tools
- Whether the product can execute multi-step work rather than only generate a one-off answer.
- Support for tool calling, file operations, browsers, or external services.
- Memory, state management, human approval, and failure-recovery capabilities.
- Clear information about model support, permissions, deployment, and data handling.
Browse by category
Published AI agent tools
The directory is sorted by quality signals and information completeness. Open a profile for sources and limitations.
Enterprise-grade distributed service mesh for MCP agents. Distributed Dynamic Dependency Injection (DDDI) with zero-boilerplate, Kubernetes-native deployment across Python, TypeScript, and Java.
Sapphire Blue AI's core capabilities include Features All autonomy coding help info personality tools voice web Personas Personas combine prompts, tools, TTS and memory slots into one persona as a preset, Toolsets Tell your AI which tools it is able to use and which are invisible, Self-Edit The AI can edit its own prompt, name, tts voice, Toolmaker The AI can make tools to interact with the world around you.
Tabnine is the AI code assistant that accelerates and simplifies software development while keeping your code private, secure, and compliant.
Run open-source machine learning models with a cloud API
Use Claude Code, Codex, OpenCode, and Pi Agent in Sublime Text through a native editor interface with project context, sessions, and tool approvals.
Annotate anything on your localhost and let your AI coding agent implement every fix in one batch. Works with Claude Code, Cursor, Windsurf, GitHub Copilot. Free, source-available, 100% local.
Agents-Flex is an open-source Java AI framework designed to simplify the development of AI applications and agents. It provides advanced features like RAG (Retrieval-Augmented Generation) to help developers build intelligent applications quickly.
ArcKit AI harness example projects: AI-generated architecture documentation for health, cloud, digital services, governance, procurement, data, and UK public-sector scenarios.
CONTINUUM is an open-source, framework-agnostic semantic recovery layer for long-running AI agents. It provides crash recovery, state verification, and full replay via 9 MCP tools and 14 CLI commands, ensuring agents don't lose state after a crash.
OptimNow is a smart cloud FinOps platform that helps you optimize cloud resources and save costs. It offers cost allocation, AI cost management, and real-time governance for AWS, Azure, and Google Cloud.
Build agents humans want to talk to. Combine the best STT, LLM, and TTS models on co-located infrastructure for ultra-low latency and production-scale reliability.
Track, visualize, and compete on AI coding assistant token usage across Claude Code, Cursor, OpenCode, Codex, Gemini, Kimi, and Qwen. The Kardashev Scale for AI Devs.
tRPC-Agent-Go is a Go agent framework for building intelligent AI applications.
The Governed Workspace Standard for AI Agents. A predictable, auditable, and agent-friendly environment.
RSIHub is an open-source framework for self-evolving agents by Simple Agent Lab: evaluator-driven learning, reproducible candidate lineage, and controllable modification. Build agents that improve and keep the evidence.
One hub for every AI coding agent. Unified configuration management for 22+ AI coding assistants. Manage MCP servers, portable skills, and project configs from a single desktop app.
AI Novel Writing Workbench is an AI-native open-source production system for long-form novels, integrating auto-director, worldbuilding, characters, chapter splitting, chapter execution, and quality repair into a pauseable and resumable production chain, helping beginners turn ideas into complete books.
Retentioneering — clickstream and user-path analysis.
StackQL is an open-source infrastructure-as-code tool and the substrate for AI agents to communicate with cloud control planes and data planes using SQL.
Frona is an open-source personal AI assistant focused on security and sandboxing, offering agent management, network control, and more.
OpenOutreach is an open-source AI lead finder that helps you describe your product and define your target market. The AI finds people who fit and tells you why each one does. It discovers people from licensed B2B data, qualifies each against your ICP, and exports them as CSV. Open source, self-host free. Send with whatever you already use.
TEN Agent is a multi-purpose voice assistant agent example powered by TEN. It demonstrates how to build and deploy voice-interactive agents for various scenarios.
Abu is a free, open-source AI desktop assistant that performs tasks locally, such as reading files, running commands, writing documents, and building reports.
The Acharya Technology offers a suite of tools for extracting, converting, and managing marker metadata from Final Cut Pro, with options to upload data to Airtable.
KimYx（老金），15年游戏行业老兵，音舞赛道深耕者，游戏研发公司联合创始人。2026 全面投入 AIGC 实践，一个人 + AI = 一支完整团队。作品集、AI Agent 实时办公室、AIGC 课程、在线工具。
FAQ
AI agent tool FAQ
What is the difference between an AI agent and a chat tool?
Agents typically plan and execute multiple steps, call tools, or operate external systems. Chat tools focus more on generating content through conversation, and exact capabilities vary by product.
Should I choose an agent product or an agent framework?
Start with a product when you need to complete work immediately. Compare frameworks or developer tools when you need to embed the workflow in your own application and control models or permissions.
How does Mengbi classify agent tools?
Classification uses public product pages, documentation, and positioning signals together with automated quality checks. Profiles retain source clues and verification information.