Self-hostable workflow automation with AI pieces and MCP access.
Tools that pair well with Claude
Editorial picks and disclosed partners. Free, open-source community resources live in the directory.
Build Python multi-agent systems with AG2, the open-source AgentOS evolved from AutoGen, including conversable agents, group chats, swarms, tools, human review, RAG, and code execution.
Monitor local Claude Code and Codex sessions from a native desktop companion and get notified when an agent needs your attention.
Open-source prompt management, evaluation, and observability for LLM apps.
Open-source observability for AI agent traces, replays, and costs.
Build and serve Python agents with event streams, permissions, workspaces, middleware, MCP, Mem0 memory, and agent teams.
Build and run agent platforms with agents, teams, workflows, memory, MCP, and AgentOS.
Terminal AI coding assistant for Git repositories.
Run a local-first AI workspace with document chat, RAG, agents, MCP-compatible tools, model routing, memories, scheduled tasks, and Docker self-hosting.
Author, schedule, monitor, and operate workflow DAGs as code.
Web automation and scraping platform.
Authenticated tool-calling platform for agents.
YAML workflow engine for deterministic, repeatable AI coding agent development.
Open-source LLM observability and evaluation tooling.
Record, replay, stream, and embed terminal sessions as text-based casts.
Fast AST-based structural search, linting, and rewriting for codebases.
Lightweight modular Python framework for building agentic AI pipelines from atomic components.
Local credential broker for AI agents.
Config-driven framework for LLM fine-tuning and post-training across full, LoRA/QLoRA, preference-tuning, and RL methods.
Open-source language for typed LLM functions with structured I/O and generated clients for many languages.
Build, package, serve, containerize, and deploy AI model inference APIs.
Browser AI app builder for web projects.
Evaluation and logging platform for AI applications.
Browser Use's editable CDP harness for coding agents that need direct control of real Chrome, screenshots, cloud browsers, and domain skills.
Open-source browser automation for AI agents.
Cloud browser infrastructure for AI agents.
Headless browser infrastructure for Puppeteer, Playwright, and AI agents.
Python multi-agent framework for agent societies, ChatAgent workflows, RAG, tool use, MCP examples, data generation, and large-scale agent research.
Use Casdoor as an IAM-backed agent and MCP gateway: manage identity, applications, users, OAuth/OIDC, SSO, MFA, and MCP server proxying from one open-source control plane.
Local CLI for coding-agent token usage and cost reports.
Build Python conversational AI apps with chat UI, steps, auth, and integrations.
Use Cherry Studio as a desktop AI client with many model providers, local models, assistants, document handling, WebDAV backup, and MCP support.
Store embeddings and metadata for AI retrieval with local, self-hosted, or cloud Chroma.
Terminal coding agent for Claude-powered development.
Find, inspect, install, publish, and update OpenClaw skills and plugins with ClawHub's hosted registry, OpenClaw commands, and npm CLI.
Open-source autonomous coding agent for VS Code.
Framework for building AI agents on Cloudflare Workers.
AI code review platform for pull requests.
Open-source AI memory platform that builds a knowledge graph for agent long-term memory.
Tool integration platform for AI agents.
Open-source AI coding assistant with model control.
Build agent-native apps with CopilotKit: generative UI, shared state, human-in-the-loop flows, AG-UI, React, mobile, chat, and agent skills.
Sign, verify, and attest containers, binaries, SBOMs, and OCI artifacts.
Open-source LLM-friendly web crawler that turns pages into clean Markdown for RAG and agents.
Multi-agent workflow framework and platform.
Terminal-based AI coding agent from Charm with multi-model support, LSP and MCP context, and permission prompts before running tools.
Build and benchmark computer-use agents with Cua drivers, MCP, sandboxes, Python SDKs, GUI automation skills, and desktop virtualization.
Codebase-aware AI editor for agent-assisted development.
Orchestrate data assets, pipelines, jobs, schedules, sensors, and observability.
Open-source sandbox infrastructure for executing AI-generated code in isolated environments, with SDKs, snapshots, and an optional managed cloud.
Transform warehouse data with SQL models, tests, docs, lineage, and artifacts.
Python LLM evaluation tests, metrics, regression checks, and tracing.
Run a ByteDance super-agent harness with skills, memory, subagents, sandboxes, MCP server support, messaging channels, and LangGraph workflows.
AI software engineering agent for development tasks.
Build agentic LLM apps with Dify's visual workflows, RAG pipelines, agents, model management, observability, prompt IDE, APIs, cloud option, and self-hosted deployment.
Open-source document parser that converts PDF, Office, HTML, and images to Markdown/JSON for gen AI.
Program and optimize LM systems with signatures, modules, tools, and metrics.
Run embedded analytical SQL over local files, data frames, and DuckDB databases.
Version datasets, models, ML pipelines, experiments, metrics, and remotes with Git.
Open-source cloud sandboxes for executing AI-generated code, with Python and JavaScript SDKs and a Code Interpreter package.
Evaluate, test, and monitor ML models, LLM apps, data quality, and drift.
Search and retrieval API for AI agents.
Web crawling API for LLM-ready content.
Visual builder for AI agents, RAG apps, chatbots, agentic workflows, multi-agent systems, LangChain components, and self-hosted flow deployment.
Open-source LLM vulnerability scanner.
Use Google's Gemini CLI as a terminal AI agent for codebase work, shell commands, web fetching, Search grounding, MCP integrations, checkpointing, and scripted automation.
AI testing platform for LLM and ML quality.
AI developer assistant across GitHub and editors.
Open-source secret scanner for repositories and files.
Build, run, evaluate, and deploy code-first AI agents and workflows.
Use `gws` to let humans and AI agents operate Google Workspace APIs through structured JSON, OAuth, helper commands, and bundled Agent Skills.
Open-source, extensible AI agent that installs, executes, edits, and tests with any LLM, as a desktop app, CLI, and API.
Cache LLM responses by semantic similarity to reduce API cost and latency across chat and agent apps.
Build and share Python machine-learning demos, AI apps, and chatbot interfaces.
AI code review assistant inside Graphite workflows.
Define, run, document, and automate data quality validations with GX Core.
Modular Python framework for AI agents and LLM workflows with tools, memory, and RAG.
Scan images, directories, SBOMs, PURLs, and CPEs for known vulnerabilities.
Open-source input/output guardrails and validators for LLM apps.
Build agents, RAG pipelines, search, retrieval, and tool-using LLM apps.
LLM observability and cost tracking platform.
Run a Nous Research agent with skills, memory, MCP integration, messaging gateway, scheduled automations, subagents, and OpenClaw migration.
Run raw PyTorch training and inference across distributed and mixed-precision setups.
Load, stream, inspect, and preprocess AI datasets for training and evaluation.
Run and train diffusion pipelines for image, video, and audio generation.
Load metrics, comparisons, and measurements for reproducible model and dataset evaluation.
Fine-tune large models with lightweight adapters, LoRA, and PEFT methods.
Build lightweight Python agents with Hugging Face Smolagents, including CodeAgent, ToolCallingAgent, MCP tools, Hub tools, LiteLLM providers, local models, CLI agents, and optional sandboxes.
Use pretrained models for inference, generation, multimodal tasks, and training.
Open-source IDE for orchestrating AI coding agents, built on Claude Code, with parallel sessions, worktrees, and team context engineering.
Cloud-hosted Chrome sessions with scraping APIs and agent integrations for Claude, OpenAI, and BrowserUse.
Structured LLM outputs with Pydantic models, validation, and retries.
Kotlin/JVM AI agent framework from JetBrains with MCP, graph workflows, memory, streaming, tools, OpenTelemetry, and Spring/Ktor integration.
Scan Kubernetes clusters, manifests, charts, repositories, and images for security risk.
Open-source data labeling and human-in-the-loop AI evaluation.
AI security guardrails for LLM applications.
Open-source AI-agent observability with tracing, evals, signals, SQL dashboards, and datasets.
Store multimodal data and run vector, full-text, and SQL retrieval for AI applications.
Build LLM apps and agents with LangChain's model interface, tools, middleware, RAG, streaming, memory, MCP adapters, LangGraph execution, and LangSmith tracing.
Java/JVM LLM framework for agents, RAG, tool calling, MCP tool providers, vector stores, chat memory, and enterprise Java integrations.
Visual builder for agents, workflows, RAG apps, and MCP-enabled LLM tools.
Open-source LLM tracing and evaluation platform.
Stateful orchestration framework for LLM agents.
Observability and evaluation platform for LLM apps.
Build stateful agents with Letta, including long-term memory, Letta Code, hosted API agents, Python and TypeScript clients, skills, subagents, custom tools, and persistent agent state.
Self-hosted ChatGPT-style platform with agents, MCP tools, Skills, Subagents, Code Interpreter, web search, artifacts, multi-provider routing, and auth.
Open-source AI gateway for routing LLM calls across providers.
Build custom AI model inference APIs with a LitAPI class, batching, streaming, and multi-GPU autoscaling.
Realtime voice and multimodal agent framework with STT, LLM, TTS, telephony, MCP, testing, and LiveKit deployment support.
Run GGUF models locally with C/C++ inference and an OpenAI-compatible server.
Build agents, RAG pipelines, retrieval, indexing, and data-aware LLM apps.
Self-hostable agent workspace for building, scheduling, grouping, and running AI agents with memory, model routing, skills, MCP-compatible plugins, and Docker deployment.
Open-source self-hosted AI engine with drop-in OpenAI/Anthropic API compatibility, no GPU required.
AI app builder for prompt-driven web products.
Visual automation platform for integrations and AI workflows.
Build reactive Python notebooks that run as scripts, apps, and git-friendly data workflows.
Python framework for structured outputs and agentic workflows with tasks, agents, and threads.
TypeScript framework for agents and AI workflows.
Visual MCP server testing and debugging.
Self-host many MCP servers as HTTP endpoints via supergateway.
Build MCP-native Python agents with composable workflow patterns, managed MCP server connections, durable execution, and agent-as-MCP-server deployment.
Build MCP servers and clients in Go with a high-level SDK covering tools, resources, prompts, transports, sessions, hooks, validation, and tracing.
Build MCP servers, MCP Apps, MCP agents, and clients with mcp-use's TypeScript/Python SDKs, app scaffolding, inspector, deployment CLI, and hosted production path.
Open-source memory layer for AI agents with multi-level memory, extraction, and retrieval.
Multi-agent software-company framework for turning one-line requirements into user stories, requirements, architecture, APIs, documents, code, and agent workflows.
Build production Python and .NET agents with Microsoft Agent Framework, including workflows, orchestration, middleware, observability, Foundry hosting, A2A, MCP, and Semantic Kernel migration paths.
Open-source framework for multi-agent AI applications.
Microsoft framework for generative AI risk identification and red-team assessment.
Run scalable vector, sparse, and hybrid search for RAG and AI retrieval systems.
Minimal CLI coding agent with confirm, human, and automatic execution modes.
Trace, evaluate, monitor, and manage agents, LLM apps, prompts, and models.
Self-hostable workflow automation for AI operations.
Run a self-hosted personal AI agent across CLI, WebUI, chat apps, MCP tools, memory, automations, and long-running workspaces.
Open-source platform that runs Claude agents in isolated containers across messaging channels, with per-agent memory and vaulted credentials.
Scan AI agent skills before installing them with NVIDIA SkillSpector: static checks, optional LLM review, MCP risk analyzers, OSV lookups, and SARIF.
Build official MCP clients and servers in .NET with the ModelContextProtocol packages for core APIs, hosting, dependency injection, and ASP.NET Core HTTP transports.
Build official MCP clients and servers in Go with the modelcontextprotocol Go SDK, including MCP, JSON-RPC, OAuth, transports, examples, and spec compatibility.
Build MCP clients and servers in Java with the official Java SDK, Maven artifacts, Java 17+, Reactive Streams, JDK HttpClient, Servlet transport, and Spring AI integration guidance.
Build MCP clients and servers in Kotlin with the official Kotlin Multiplatform SDK, Gradle artifacts, Ktor integration, coroutine APIs, and cross-platform transports.
Build MCP clients and servers in PHP with the official `mcp/sdk` Composer package, framework-agnostic APIs, attributes, stdio and HTTP transports, and session support.
Build MCP clients and servers in Python with the official `mcp` SDK package, including FastMCP, tools, resources, prompts, stdio, SSE, Streamable HTTP, authentication, and development tooling.
Build MCP clients and servers in Ruby with the official `mcp` gem, JSON-RPC, tools, resources, prompts, stdio, Streamable HTTP, Rack, and Rails integration.
Build MCP clients and servers in Rust with the official `rmcp` SDK, tokio async runtime support, tool macros, resources, prompts, transports, and docs.rs API documentation.
Build MCP clients and servers in Swift with the official Swift SDK, Swift Package Manager, Apple platform support, stdio and HTTP transports, and MCP client/server APIs.
Build MCP clients and servers in TypeScript with the official modelcontextprotocol SDK, including stdio and HTTP transports, tools, resources, prompts, OAuth helpers, and examples.
Local model runner for open models and developer workflows.
Self-hosted coding agent framework built on LangGraph — trigger via Slack, Linear, or GitHub and get automatic PRs from isolated sandboxes.
Self-hosted AI web UI for Ollama and OpenAI-compatible APIs with RAG, tools, model builder, web search, vector databases, plugins, and enterprise auth.
Build JavaScript and TypeScript multi-agent workflows with the official OpenAI Agents SDK, including tools, handoffs, guardrails, sessions, tracing, MCP tools, realtime agents, and sandbox agents.
Build Python multi-agent workflows with the official OpenAI Agents SDK, including tools, handoffs, guardrails, sessions, tracing, realtime agents, MCP integrations, and sandbox agents.
OpenAI's open-source terminal coding agent that edits and runs code locally with sandbox and approval modes.
Build repeatable LLM and agent evals with OpenAI's open-source framework.
Generate clients, servers, docs, schemas, and config from OpenAPI specs.
Terminal-first AI coding agent for local development.
Run coding agents with a local GUI, CLI, SDK, sandboxes, and hosted options.
Sandbox runtime for AI agents with multi-language SDKs, MCP tools, Docker, Kubernetes, egress controls, credential vault, and code execution.
Python library for guaranteed structured LLM outputs (JSON schema, regex, grammar) during generation.
Build real-time voice and multimodal AI agents by composing STT, LLM, TTS, and transport services in a pipeline.
Developer workflow automation and API integration platform.
Use a fast Rust DataFrame query engine for local, lazy, and cloud-backed analytics.
Open-source AI gateway to 1600+ LLMs with retries, fallbacks, routing, guardrails, caching, and observability.
Orchestrate resilient Python data pipelines with flows, tasks, schedules, and workers.
Open-source Microsoft toolkit to build, trace, evaluate, and deploy LLM app flows.
Open-source prompt testing and red-teaming framework.
AI security platform for models and applications.
Type-safe Python agent framework with tools, outputs, MCP, and evals.
Run Qwen Code as a terminal coding agent with Auto-Memory, Auto-Skills, SubAgents, Agent Teams, MCP, multi-provider routing, IDE plugins, desktop app, daemon mode, and SDKs.
Build Qwen-powered agents with function calling, planning, memory, RAG, MCP, code interpreter, Gradio demos, BrowserQwen, and custom assistant workflows.
Open-source evaluation framework for RAG and LLM application testing.
RAGFlow is a self-hostable RAG engine and context layer with DeepDoc parsing, visual chunking, grounded citations, agent workflows, MCP, and Docker.
Scale Python and AI workloads with Ray Core and Ray AI libraries.
macOS launcher and extension platform with AI workflows.
Build full-stack web apps and AI-app UIs entirely in Python, compiled to a React and Next.js frontend.
Browser-based AI app builder inside Replit.
Open-source visual editor and TypeScript library for building and embedding LLM agent graphs.
VS Code coding agent with planning and workflow modes.
Static analysis, SAST, secrets, dependency checks, and custom code rules.
Build embeddings, sparse encoders, rerankers, and semantic search pipelines.
Manage SKILL.md agent skills with `npx skills`: add, use, find, list, update, remove, initialize, symlink, copy, and target supported agent hosts.
Keep one source of truth for AI CLI skills, agents, rules, commands, and prompts, then sync them into Codex, Claude, OpenClaw, Cursor, and more.
MCP server discovery and deployment platform.
Scan Claude, Cursor, Codex, Gemini, Windsurf, VS Code, and other local agent configs for MCP and skill supply-chain risks.
AI coding assistant for large codebase context.
Generate SDKs, CLIs, Terraform providers, and MCP servers from OpenAPI.
AI-assisted browser automation framework.
Python and TypeScript agent SDK with model providers, MCP, streaming, structured output, hooks, observability, tools, and multi-agent patterns.
Turn Python scripts into interactive data apps, dashboards, and chat interfaces.
Open-source framework that extends Claude Code with slash commands, specialized agents, behavioral modes, and MCP integrations.
Generate SBOMs from images, directories, files, archives, and OCI layouts.
Build resilient long-running workflows with durable execution and worker-backed activities.
Background jobs and workflows for TypeScript apps.
Evaluate and trace agents, RAG systems, LLM apps, metrics, and regressions.
Open-source embeddings database and AI framework for semantic search, RAG, workflows, and agents.
Open-source library for fast, low-VRAM LLM fine-tuning and RL with export to GGUF, vLLM, and Ollama.
Build Python autonomous agents with custom tools, MCP tools, prebuilt agents, workspace restrictions, sandbox execution, OCR, and document workflows.
AI interface builder for React and web UI.
TypeScript toolkit for AI apps and streaming UI.
Serve open models with high-throughput inference and OpenAI-compatible APIs.
TypeScript agent framework with workflows, MCP tools, memory, RAG, guardrails, evals, voice, and VoltOps observability.
Tracking and evaluation toolkit for LLM applications.
Build AI retrieval systems with semantic search, hybrid search, RAG, and cloud-native deployment.
Agentic coding environment for multi-file AI development.
Enterprise automation platform for business workflows.
AI-assisted automation across business apps.
Fast collaborative code editor with AI assistance.