AI Tools Library.

Open-source workflow automation with a node-based canvas and code escape hatch (JavaScript or Python). Self-hostable, which most alternatives are not.

Visual workflow builder with branching, error handling, and parallel execution. More flexible than Zapier for complex logic, easier than n8n for non-developers.

The original no-code automation tool. Largest connector library (7,000+ apps). AI Actions add LLM steps inside Zaps. Pricing scales with task volume.

API-first workflow platform where each step is Node.js or Python. Thin orchestration layer over real code. For developers who want serverless plumbing built in.

Enterprise iPaaS with governance, reusable recipes, and SLAs. Mid-market and up. Heavier price tag than Make or Zapier.

Microsoft's automation platform. Deep Microsoft 365 and Azure integration, desktop RPA, Copilot Studio for agents. Default choice in Microsoft-heavy orgs.

Enterprise-grade low-code with a Merlin AI agent layer. Strong for embedded automation (white-labelled into SaaS products).

Lightweight consumer automations. Mostly relevant for personal and smart-home workflows now, not B2B.

Visual AI workflow builder focused on web research, scraping, and data extraction. LLM steps are first-class primitives, not afterthoughts.

AI workflow tool with human-in-the-loop built in. Approvals, forms, and review steps as first-class features. Differentiates from autonomous agents.

AI employees with defined roles. Faster path to running agents than building workflows. Strong for sales and customer success use cases.

Browser-native automation extension. Automates repetitive GTM and admin work directly in the apps you already use (LinkedIn, Gmail, CRMs, sheets).

Open-source LLM app platform with visual prompt orchestration, RAG, and an agent builder. Self-host or use cloud version.

Open-source drag-and-drop builder for LangChain flows. Visual entry point into the LangChain ecosystem without writing code directly.

AI workflow platform aimed at enterprise teams building internal AI tools without engineering bandwidth.

LLM development platform with workflow orchestration, evals, and prompt management. Popular with technical product teams.

Agentic process automation with pre-built agents for finance, ops, and customer service.

Fully managed AI automation. Their team builds and maintains workflows for you (2,500+ pre-built bots). Best when you have no internal engineering.

Direct API access to Claude models. Tool use, vision, long context, prompt caching. The building block for custom workflows that need an LLM step.

GPT models, embeddings, image generation, voice. Broadest model coverage of any single provider.

Google's Gemini models with very long context windows and native multimodal input. Strong for video and document analysis.

Single API for hundreds of models from different providers. Useful when you want to swap models without changing code, or compare outputs.

Run open-source models via API. Strong for image, video, and audio models that aren't available through OpenAI or Anthropic.

Inference platform built around custom chips. Very fast token generation for open models. Useful when latency matters.

European AI lab with both proprietary and open-source models. Strong OCR and document understanding offerings.

Inference platform for open-source models with fine-tuning support. Lower cost than the big closed providers for many workloads.

Web scraping API that returns clean Markdown ready for LLMs. Built for AI agents and workflows. Handles JavaScript-rendered pages.

Web scraping platform with thousands of pre-built scrapers ("Actors") for specific sites. Useful when you need LinkedIn, Google Maps, or similar source data.

Enterprise web data platform with massive proxy networks. Strongest option when sites actively block scrapers.

Simple scraping API with built-in proxy rotation and JavaScript rendering. Cheaper alternative to Bright Data for moderate volumes.

Agentic document parsing. Routes each page element (text, table, chart) to the best-fit model. Strong LLM-ready output for PDFs and complex documents.

High-fidelity document extraction for visually complex PDFs. Multi-pass with layout-aware analysis. SOC 2, HIPAA, on-prem options.

Open-source and enterprise platform that converts PDFs, emails, HTML, and Office files into labelled elements ready for RAG.

Open-source document parser from IBM with MCP server support. Good free option for agentic contexts.

Managed GCP service with pre-trained processors (invoices, tax forms, IDs) and custom models.

Amazon's managed forms, tables, and document extraction service. Default choice in AWS-heavy stacks.

Code-first framework for stateful agent workflows. Directed graphs with branching, retries, and human-in-the-loop checkpoints. The 2026 default for production agents.

Anthropic's SDK for building agents with native tool use, memory, and subagents. Powers Claude Code.

OpenAI's official agent SDK. Lowest barrier to entry in the OpenAI ecosystem. Production-ready as of 2026.

Multi-agent framework with role-based "crews". Each agent has a role, goal, and backstory. Fastest path to a multi-agent MVP.

Microsoft's multi-agent framework, now community-forked as AG2. Conversation-based agent orchestration.

Data-first agent framework. Best when your agent's value comes from RAG over your own documents and data.

Type-safe Python agent framework. Appeals to teams that want strong typing and validation.

Hierarchical agent trees (root delegates to sub-agents). Tight Vertex AI and Gemini integration.

Lightweight, .NET-friendly orchestration. Good for enterprise Microsoft stacks.

Market leader. Strong for enterprise RPA and AI workflows where Citrix, mainframes, or desktop steps still matter.

Attended and unattended automation at scale. AI Agent Studio for agentic workflows.

Enterprise RPA with strong governance, popular in financial services.

Enterprise AI assistant for IT, HR, and employee support. Strong in Fortune 500.

AI-powered enterprise search and workflow agents grounded in company knowledge.

Agentic platform with "AI for IT", "AI for HR", etc. Enterprise-grade with Gartner recognition.

Agent layer on top of Salesforce data and Flow automation.

Agents inside the ServiceNow workflow engine for IT service management and beyond.

AI-first code editor (fork of VS Code) with file-aware completions, Composer multi-file edits, and an in-editor agent mode. The 2026 daily-driver IDE.

Anthropic's terminal-native coding agent. Strong on multi-file refactors, codebase exploration, and long-running tasks. Pairs with any editor.

The original AI coding assistant. Lowest-friction entry point for teams new to agentic coding, deeply integrated with GitHub and VS Code.

IDE with Cascade agent and codebase indexing. Acquired by Cognition. Value alternative to Cursor at $15/month.

Open-source VS Code extension for agentic coding. Bring-your-own-key flexibility. Best for cost transparency and model independence.

Terminal-only minimalist AI pair programmer. Open-source, BYOK, integrates directly with git. Strong for command-line-first developers.

OpenAI's open-source terminal coding agent. Standalone cloud agent option with desktop apps for macOS and Windows.

Cognition's autonomous software engineer agent. Designed for delegating multi-step tasks rather than pair programming.

Claude operating a real Chrome browser, with the safety guardrails built in. Useful when an action needs a UI rather than an API.

Open-source browser automation framework for building custom web agents. Python-first.

ChatGPT-based agent that browses and acts on the web on your behalf.

Anthropic's API for letting Claude control a virtual desktop. Powerful but currently developer-only.

Voice agent infrastructure (TTS, STT, LLM orchestration) for developers. Build phone-call agents via API.

Phone-call agents for sales, support, and operations. Lower-code than Vapi for non-developers.

Real-time voice agent platform with a focus on natural conversation quality.

No-code voice agents. Build phone bots without writing any code.

ElevenLabs' conversational agent platform on top of their voice models. Strongest on voice quality.

Spreadsheet-shaped workflow tool for sales and marketing. AI columns enrich and research prospect data at scale. The reference tool for modern outbound.

Sales engagement platform with AI-assisted prospecting, sequencing, and dialer. Both a data source and an automation surface.

Sales engagement platform with embedded AI for messaging and forecasting.

AI sales agents for outbound and follow-up.

AI SDR ("Alice") and other digital workers for GTM teams.

AI BDR product ("Ava") that runs outbound campaigns autonomously.

Marketing copy and workflow platform with brand voice controls.

AI marketing platform with GTM workflows and content generation.

HubSpot's agent and AI layer across CRM, marketing, and service.

AI customer support agents for mid-market and enterprise SaaS. Strong on regulated industries.

Conversational AI for customer experience. Founded by ex-Salesforce CEO Bret Taylor.

Enterprise customer support automation platform.

Customer support agents native to the Zendesk ticket flow.

Intercom's AI agent for customer support, built into their existing platform.

Database-as-app with native AI fields and automations. Useful when a workflow needs a backing store and a UI for non-technical teammates.

Workspace with AI agents inside databases and pages. Workflows can read from and write to Notion via API.

Document-database hybrid with AI building blocks inside Coda docs.

Internal tool builder with a workflow engine. Best when you need both automation and a custom UI for employees to trigger it.

No-code apps with AI features layered on top. Build mobile-friendly internal tools from spreadsheets.

Managed vector database for embeddings. Default choice for RAG workflows that need fast similarity search at scale.

Postgres with pgvector extension built in. Combines a regular database, auth, and vector search in one platform. Good for workflow backends.

Open-source vector database with hybrid search (keyword + vector). Self-hostable.

Open-source vector database written in Rust. Strong performance and a generous free cloud tier.

Lightweight embedded vector database. Easiest option for local development and small projects.

Open-source LLM observability and prompt management. Tracing, evals, and cost tracking for AI workflows.

LangChain's hosted tracing, evals, and monitoring product. Strongest fit when already using LangChain or LangGraph.

Open-source LLM observability with caching and rate limiting.

Eval and observability platform for LLM products. Strong on evals as a first-class workflow.

Prompt management, evals, and collaboration for product teams.

Durable execution engine for long-running, fault-tolerant agent workflows. Increasingly used as the layer underneath production agents.

Event-driven durable workflow engine, AI-friendly. Lighter weight than Temporal for many use cases.

Legal AI workflow platform for law firms. Trained on legal corpora and contracts.

AI workflow tool for finance and consulting. Document-heavy research and analysis.

AI search and assistants for product documentation. Built by the Firecrawl team.

AI-powered search and chat for technical documentation.