Featured
The default model for most agent workflows in 2026.
Featured
The model to reach for when the job involves images, video, audio, or very long documents.
Featured
When you need token latency measured in milliseconds, not hundreds of milliseconds, Groq is it.
Featured
Still the default for a lot of production agents.
Featured
Useful when an agent needs to swap models without rewiring its caller.
Featured
The default when you want to run a specific open model in production without standing up your own GPU fleet.
View tool→LLM Inference
The Agent Index
The Builder Weekly · Index

The Agent Index

Infrastructure for AI agents. Tools an agent reaches for mid-workflow to do its job. Verified by the builders who use them every day.

LLM Inference(8)

  • Anthropic API

    Claude models (Sonnet, Opus, Haiku) via REST API with native tool-use, extended thinking, and MCP support.

    LLM InferenceREST API
    Visit →

  • Cerebras Inference

    OpenAI-compatible inference for open models on Cerebras' wafer-scale silicon, with sustained throughput in the thousands of tokens per second.

    LLM InferenceREST API
    Visit →

  • Google Gemini API

    Gemini 3 Pro and Flash models with native multimodal input, 2M token context, and image generation.

    LLM InferenceREST API
    Visit →

  • Groq

    Ultra-low-latency inference for open models (Llama, Mixtral, Qwen) on custom LPU silicon.

    LLM InferenceREST API
    Visit →

  • Mistral La Plateforme

    Mistral's hosted API for the Mistral Large, Codestral, and the Ministral family, with structured output, function calling, and embeddings on the same surface.

    LLM InferenceREST API
    Visit →

  • OpenAI API

    GPT-4 class models, embeddings, image generation, and the Responses API via REST.

    LLM InferenceREST API
    Visit →

  • OpenRouter

    OpenAI-compatible router that exposes hundreds of frontier and open models behind one API key, one URL, and one pricing surface.

    LLM InferenceREST API
    Visit →

  • Together AI

    Hosted inference for hundreds of open models plus dedicated endpoints and fine-tuning.

    LLM InferenceREST API
    Visit →
Memory(4)

  • Cognee

    Open-source memory layer that builds a queryable knowledge graph from an agent's accumulated context, replacing flat vector retrieval with structured semantic recall.

    MemoryPython Library
    Visit →GitHub →

  • Letta

    Stateful agent platform with persistent memory, context management, and a REST API for agent servers.

    MemoryREST API
    Visit →GitHub →

  • Mem0

    Persistent memory layer for AI agents with user-scoped storage, semantic recall, and fact extraction.

    MemoryREST API
    Visit →GitHub →

  • Zep

    Long-term memory for AI agents built on a temporal knowledge graph (Graphiti).

    MemoryPython Library
    Visit →GitHub →
Vector Stores(5)

  • Chroma

    Embedded and hosted vector database with a simple Python-first API, popular for RAG prototyping.

    Vector StoresPython Library
    Visit →GitHub →

  • Pinecone

    Fully managed vector database with serverless indexes, hybrid search, and metadata filtering.

    Vector StoresREST API
    Visit →

  • Qdrant

    Rust-based vector search engine with strong filtering, sparse vector support, and a managed cloud.

    Vector StoresREST API
    Visit →GitHub →

  • Turbopuffer

    Object-storage-native vector database designed to store billions of vectors cheaply on S3.

    Vector StoresREST API
    Visit →

  • Weaviate

    Open-source vector database with native hybrid search, multi-tenancy, and generative modules.

    Vector StoresREST API
    Visit →GitHub →
Voice and Speech(5)

  • AssemblyAI

    Speech intelligence API with transcription, speaker diarization, sentiment, and topic extraction.

    Voice and SpeechREST API
    Visit →

  • Cartesia

    State-space-model TTS with sub-100ms first-token latency and streaming voice synthesis.

    Voice and SpeechREST API
    Visit →

  • Deepgram

    Speech-to-text and text-to-speech APIs tuned for low latency and real-time agent pipelines.

    Voice and SpeechREST API
    Visit →

  • ElevenLabs

    High-fidelity voice synthesis with voice cloning, streaming TTS, and a conversational agent API.

    Voice and SpeechREST API
    Visit →

  • Vapi

    A voice agent runtime that wires STT, LLM, and TTS together behind a single API. Define an assistant, attach tools, attach a system prompt, and Vapi handles the WebRTC session, turn-taking, and barge-in.

    Voice and SpeechREST API
    Visit →
Browser Automation(5)

  • Browserbase

    Managed headless browser infrastructure for AI agents with session recording, live view, and stealth.

    Browser AutomationSDK
    Visit →

  • Hyperbrowser

    Managed Chromium sessions an agent drives over CDP, with built-in proxy rotation, CAPTCHA handling, and session persistence.

    Browser AutomationREST API
    Visit →

  • Skyvern

    Open-source browser automation where the LLM reads the rendered page and decides the next click instead of running a brittle Playwright script.

    Browser AutomationREST API
    Visit →

  • Stagehand

    High-level browser automation library that lets agents describe actions in natural language.

    Browser AutomationTypeScript Library
    Visit →GitHub →

  • Steel Browser

    Open-source browser API for AI agents with session persistence, proxies, and CAPTCHA handling.

    Browser AutomationREST API
    Visit →GitHub →
Code Execution(4)

  • Daytona

    Open-source development environments on demand for AI agents to read, edit, and execute codebases.

    Code ExecutionCLI
    Visit →GitHub →

  • E2B

    Secure cloud sandboxes for AI agents to run code, render files, and execute long-running processes.

    Code ExecutionSDK
    Visit →GitHub →

  • Modal

    Serverless cloud platform for running Python functions, GPU jobs, and containers on demand.

    Code ExecutionPython Library
    Visit →

  • Riza

    Secure code interpreter API for LLM-generated code with fast-start sandboxes and memory isolation.

    Code ExecutionREST API
    Visit →
Web Search and Retrieval(5)

  • Brave Search API

    Search API backed by Brave's own crawled index, not a Bing or Google reseller. Three tiers: free for low-volume agents, pro for higher throughput, AI tier for embed and re-ranking heavy workflows.

    Web Search and RetrievalREST API
    Visit →

  • Exa

    Neural web search API designed for LLMs, with semantic matching, live-crawled content, and similarity queries.

    Web Search and RetrievalREST API
    Visit →

  • Firecrawl

    Scrape any website into clean markdown or structured data with a single API call, including JS-rendered pages.

    Web Search and RetrievalREST API
    Visit →GitHub →

  • Linkup

    Search API designed for agent consumption, returning ranked results with deep-content extraction and source attribution from the live web.

    Web Search and RetrievalREST API
    Visit →

  • Tavily

    Search API built specifically for AI agents with grounded answers, source citations, and news search.

    Web Search and RetrievalREST API
    Visit →
Scheduling and Orchestration(4)

  • CueAPI

    Scheduling and execution accountability API for AI agents with verified outcomes and retries.

    Scheduling and OrchestrationREST API
    Visit →GitHub →

  • Hatchet

    Durable execution engine that runs workflows as code, with native step retries, fan-out/fan-in, and a typed step API. Self-host in Docker or run as Hatchet Cloud.

    Scheduling and OrchestrationSDK
    Visit →GitHub →

  • Inngest

    Durable execution platform for background jobs, AI workflows, and long-running agent steps.

    Scheduling and OrchestrationSDK
    Visit →GitHub →

  • Trigger.dev

    Background job platform with built-in AI task support, concurrency controls, and wait-until-event primitives.

    Scheduling and OrchestrationSDK
    Visit →GitHub →
Communication(3)

  • Postmark

    Transactional email API focused on inbox placement, not marketing volume. Fast bounce/spam reporting and a sandbox tier that lets agents test without a verified domain.

    CommunicationREST API
    Visit →

  • Resend

    Developer-first transactional email API with React Email integration and high deliverability.

    CommunicationREST API
    Visit →

  • SendGrid

    Email delivery infrastructure for transactional and marketing mail at high volume.

    CommunicationREST API
    Visit →
Credentials and Tool Management(3)

  • Arcade AI

    Authenticated tool-calling platform for agents with permissioned actions and per-user credential storage.

    Credentials and Tool ManagementSDK
    Visit →

  • Composio

    Tool and credential management platform for AI agents with 250+ pre-built integrations and OAuth handling.

    Credentials and Tool ManagementSDK
    Visit →GitHub →

  • Metorial

    Managed MCP and tool-use platform for AI agents with credential vaulting and usage analytics.

    Credentials and Tool ManagementMCP Server
    Visit →
MCP Ecosystem(3)

  • Context7

    MCP server that feeds agents up-to-date library documentation and code examples from the source.

    MCP EcosystemMCP Server
    Visit →GitHub →

  • Model Context Protocol

    Open standard for connecting AI agents to tools and data sources, with a reference server collection.

    MCP EcosystemMCP Server
    Visit →GitHub →

  • Smithery

    MCP server registry and hosting with one-click install for Claude, Cursor, and other MCP clients.

    MCP EcosystemMCP Server
    Visit →
Guardrails and Safety(2)

  • Guardrails AI

    Open-source framework for adding runtime validators around LLM input and output, with a Hub of pre-built validators for PII detection, jailbreak attempts, hallucination checks, and structured-output enforcement.

    Guardrails and SafetyPython Library
    Visit →GitHub →

  • Lakera Guard

    A scanning API an agent calls before passing user input to the model, and again before passing model output to a downstream tool. Trained on a large corpus of jailbreaks, prompt-injection patterns, and PII templates, with low-latency calls that don't break the agent's loop.

    Guardrails and SafetyREST API
    Visit →
Add to the list

The Agent Index is open source. The canonical list lives on GitHub, and every entry on this page is generated from that repo at build time. Contributions are welcome as pull requests. Read the rules before submitting; the bar is sharp, but the review is fast.

Submit on GitHub →Read the contribution rules
How this list gets made

Tools are added after they show up in production agent workflows or get strong signal from builders with real usage. Inclusion is the editorial judgment. There is no ranking within categories. Every tool listed has earned its place against the same criterion: can an agent complete its full lifecycle through a fully programmatic path?

How we choose tools →